linux-doc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Xin Li <xin@zytor.com>
To: Chao Gao <chao.gao@intel.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-doc@vger.kernel.org, seanjc@google.com,
	pbonzini@redhat.com, corbet@lwn.net, tglx@linutronix.de,
	mingo@redhat.com, bp@alien8.de, dave.hansen@linux.intel.com,
	x86@kernel.org, hpa@zytor.com, luto@kernel.org,
	peterz@infradead.org, andrew.cooper3@citrix.com
Subject: Re: [PATCH v3 25/27] KVM: nVMX: Add FRED VMCS fields
Date: Fri, 25 Oct 2024 00:25:45 -0700	[thread overview]
Message-ID: <f9bb0740-21ec-482d-92fb-7fed3fef7d36@zytor.com> (raw)
In-Reply-To: <Zxn6Vc/2vvJ3VHCb@intel.com>

On 10/24/2024 12:42 AM, Chao Gao wrote:
>> @@ -7197,6 +7250,9 @@ static void nested_vmx_setup_basic(struct nested_vmx_msrs *msrs)
>> 	msrs->basic |= VMX_BASIC_TRUE_CTLS;
>> 	if (cpu_has_vmx_basic_inout())
>> 		msrs->basic |= VMX_BASIC_INOUT;
>> +
>> +	if (cpu_has_vmx_fred())
>> +		msrs->basic |= VMX_BASIC_NESTED_EXCEPTION;
> 
> why not advertising VMX_BASIC_NESTED_EXCEPTION if the CPU supports it? just like
> VMX_BASIC_INOUT right above.

Because VMX nested-exception support only works with FRED.

We could pass host MSR_IA32_VMX_BASIC.VMX_BASIC_NESTED_EXCEPTION to
nested, but it's meaningless w/o VMX FRED.

> 
> 
>> }
>>
>> static void nested_vmx_setup_cr_fixed(struct nested_vmx_msrs *msrs)
>> diff --git a/arch/x86/kvm/vmx/nested.h b/arch/x86/kvm/vmx/nested.h
>> index 2c296b6abb8c..5272f617fcef 100644
>> --- a/arch/x86/kvm/vmx/nested.h
>> +++ b/arch/x86/kvm/vmx/nested.h
>> @@ -251,6 +251,14 @@ static inline bool nested_cpu_has_encls_exit(struct vmcs12 *vmcs12)
>> 	return nested_cpu_has2(vmcs12, SECONDARY_EXEC_ENCLS_EXITING);
>> }
>>
>> +static inline bool nested_cpu_has_fred(struct vmcs12 *vmcs12)
>> +{
>> +	return vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_FRED &&
>> +	       vmcs12->vm_exit_controls & VM_EXIT_ACTIVATE_SECONDARY_CONTROLS &&
>> +	       vmcs12->secondary_vm_exit_controls & SECONDARY_VM_EXIT_SAVE_IA32_FRED &&
>> +	       vmcs12->secondary_vm_exit_controls & SECONDARY_VM_EXIT_LOAD_IA32_FRED;
> 
> Is it a requirement in the SDM that the VMM should enable all FRED controls or
> none? If not, the VMM is allowed to enable only one or two of them. This means
> KVM would need to emulate FRED controls for the L1 VMM as three separate
> features.

The SDM doesn't say that.  But FRED states are used during and
immediately after VM entry and exit, I don't see a good reason for a VMM
to enable only one or two of the 3 save/load configs.

Say if VM_ENTRY_LOAD_IA32_FRED is not set, it means a VMM needs to
switch to guest FRED states before it does a VM entry, which is
absolutely a big mess.

TBH I'm not sure this is the question you have in mind.

  reply	other threads:[~2024-10-25  7:26 UTC|newest]

Thread overview: 81+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-01  5:00 [PATCH v3 00/27] Enable FRED with KVM VMX Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 01/27] KVM: x86: Use a dedicated flow for queueing re-injected exceptions Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 02/27] KVM: VMX: Don't modify guest XFD_ERR if CR0.TS=1 Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 03/27] KVM: VMX: Add support for the secondary VM exit controls Xin Li (Intel)
2024-10-21  8:28   ` Chao Gao
2024-10-21 17:03     ` Xin Li
2024-10-22  2:47       ` Chao Gao
2024-10-22 16:30         ` Xin Li
2025-02-25 17:28           ` Sean Christopherson
2024-10-01  5:00 ` [PATCH v3 04/27] KVM: VMX: Initialize FRED VM entry/exit controls in vmcs_config Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 05/27] KVM: VMX: Disable FRED if FRED consistency checks fail Xin Li (Intel)
2024-10-22  8:48   ` Chao Gao
2024-10-22 16:21     ` Xin Li
2024-11-26 15:32   ` Borislav Petkov
2024-11-26 18:53     ` Xin Li
2024-11-26 19:04       ` Borislav Petkov
2024-10-01  5:00 ` [PATCH v3 06/27] x86/cea: Export per CPU variable cea_exception_stacks Xin Li (Intel)
2024-10-01 16:12   ` Dave Hansen
2024-10-01 17:51     ` Xin Li
2024-10-01 18:18       ` Dave Hansen
2024-10-01  5:00 ` [PATCH v3 07/27] KVM: VMX: Initialize VMCS FRED fields Xin Li (Intel)
2024-10-22  9:06   ` Chao Gao
2024-10-22 16:18     ` Xin Li
2024-10-01  5:00 ` [PATCH v3 08/27] KVM: x86: Use KVM-governed feature framework to track "FRED enabled" Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 09/27] KVM: VMX: Do not use MAX_POSSIBLE_PASSTHROUGH_MSRS in array definition Xin Li (Intel)
2024-11-26 18:02   ` Borislav Petkov
2024-11-26 19:22     ` Xin Li
2024-11-26 20:06       ` Borislav Petkov
2024-11-27  6:46         ` Xin Li
2024-11-27  6:55           ` Borislav Petkov
2024-11-27  7:02             ` Xin Li
2024-11-27  7:10               ` Borislav Petkov
2024-11-27  7:32                 ` Xin Li
2024-11-27  7:58                   ` Borislav Petkov
2024-10-01  5:00 ` [PATCH v3 10/27] KVM: VMX: Set FRED MSR interception Xin Li (Intel)
2024-11-13 11:31   ` Chao Gao
2024-10-01  5:00 ` [PATCH v3 11/27] KVM: VMX: Save/restore guest FRED RSP0 Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 12/27] KVM: VMX: Add support for FRED context save/restore Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 13/27] KVM: x86: Add a helper to detect if FRED is enabled for a vCPU Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 14/27] KVM: VMX: Pass XFD_ERR as pseudo-payload when injecting #NM Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 15/27] KVM: VMX: Virtualize FRED event_data Xin Li (Intel)
2024-10-01  5:00 ` [PATCH v3 16/27] KVM: VMX: Virtualize FRED nested exception tracking Xin Li (Intel)
2024-10-24  6:24   ` Chao Gao
2024-10-25  8:04     ` Xin Li
2024-10-28  6:33       ` Chao Gao
2024-12-05  7:16         ` Xin Li
2024-10-01  5:01 ` [PATCH v3 17/27] KVM: x86: Mark CR4.FRED as not reserved when guest can use FRED Xin Li (Intel)
2024-10-24  7:18   ` Chao Gao
2024-12-12 18:48     ` Xin Li
2024-12-12 19:05       ` Sean Christopherson
2024-12-13 18:43         ` Xin Li
2024-10-01  5:01 ` [PATCH v3 18/27] KVM: VMX: Dump FRED context in dump_vmcs() Xin Li (Intel)
2024-10-24  7:23   ` Chao Gao
2024-10-24 16:50     ` Xin Li
2024-10-01  5:01 ` [PATCH v3 19/27] KVM: x86: Allow FRED/LKGS to be advertised to guests Xin Li (Intel)
2024-10-01  5:01 ` [PATCH v3 20/27] KVM: x86: Allow WRMSRNS " Xin Li (Intel)
2025-02-25 15:41   ` Sean Christopherson
2024-10-01  5:01 ` [PATCH v3 21/27] KVM: VMX: Invoke vmx_set_cpu_caps() before nested setup Xin Li (Intel)
2024-10-24  7:49   ` Chao Gao
2024-10-25  7:34     ` Xin Li
2025-02-25 16:01       ` Sean Christopherson
2024-10-01  5:01 ` [PATCH v3 22/27] KVM: nVMX: Add support for the secondary VM exit controls Xin Li (Intel)
2024-10-01  5:01 ` [PATCH v3 23/27] KVM: nVMX: Add a prerequisite to SHADOW_FIELD_R[OW] macros Xin Li (Intel)
2024-10-01  5:01 ` [PATCH v3 24/27] KVM: nVMX: Add a prerequisite to existence of VMCS fields Xin Li (Intel)
2025-02-25 16:22   ` Sean Christopherson
2025-02-25 16:37     ` Xin Li
2025-02-25 19:32       ` Sean Christopherson
2024-10-01  5:01 ` [PATCH v3 25/27] KVM: nVMX: Add FRED " Xin Li (Intel)
2024-10-24  7:42   ` Chao Gao
2024-10-25  7:25     ` Xin Li [this message]
2024-10-28  9:07       ` Chao Gao
2024-10-28 18:27         ` Sean Christopherson
2024-10-29 17:40           ` Xin Li
2024-10-01  5:01 ` [PATCH v3 26/27] KVM: nVMX: Add VMCS FRED states checking Xin Li (Intel)
2024-10-01  5:01 ` [PATCH v3 27/27] KVM: nVMX: Allow VMX FRED controls Xin Li (Intel)
2025-02-19  0:26 ` [PATCH v3 00/27] Enable FRED with KVM VMX Xin Li
2025-02-25 15:24   ` Sean Christopherson
2025-02-25 17:04     ` Xin Li
2025-02-25 17:35       ` Sean Christopherson
2025-02-25 18:48         ` Xin Li
2025-02-28 17:06 ` Sean Christopherson

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=f9bb0740-21ec-482d-92fb-7fed3fef7d36@zytor.com \
    --to=xin@zytor.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=bp@alien8.de \
    --cc=chao.gao@intel.com \
    --cc=corbet@lwn.net \
    --cc=dave.hansen@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=mingo@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=seanjc@google.com \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /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).