From: Alexey Kardashevskiy <aik@amd.com>
To: Sean Christopherson <seanjc@google.com>
Cc: Dionna Amalie Glaze <dionnaglaze@google.com>,
Michael Roth <michael.roth@amd.com>,
kvm@vger.kernel.org, linux-coco@lists.linux.dev,
linux-mm@kvack.org, linux-crypto@vger.kernel.org, x86@kernel.org,
linux-kernel@vger.kernel.org, tglx@linutronix.de,
mingo@redhat.com, jroedel@suse.de, thomas.lendacky@amd.com,
hpa@zytor.com, ardb@kernel.org, pbonzini@redhat.com,
vkuznets@redhat.com, jmattson@google.com, luto@kernel.org,
dave.hansen@linux.intel.com, slp@redhat.com, pgonda@google.com,
peterz@infradead.org, srinivas.pandruvada@linux.intel.com,
rientjes@google.com, dovmurik@linux.ibm.com, tobin@ibm.com,
bp@alien8.de, vbabka@suse.cz, kirill@shutemov.name,
ak@linux.intel.com, tony.luck@intel.com, marcorr@google.com,
sathyanarayanan.kuppuswamy@linux.intel.com, alpergun@google.com,
jarkko@kernel.org, ashish.kalra@amd.com, nikunj.dadhania@amd.com,
pankaj.gupta@amd.com, liam.merwick@oracle.com,
zhi.a.wang@intel.com, Brijesh Singh <brijesh.singh@amd.com>
Subject: Re: [PATCH v10 48/50] KVM: SEV: Provide support for SNP_GUEST_REQUEST NAE event
Date: Thu, 19 Oct 2023 13:48:26 +1100 [thread overview]
Message-ID: <924b755a-977a-4476-9525-a7626d728e18@amd.com> (raw)
In-Reply-To: <ZS_iS4UOgBbssp7Z@google.com>
On 19/10/23 00:48, Sean Christopherson wrote:
> On Wed, Oct 18, 2023, Alexey Kardashevskiy wrote:
>>
>> On 18/10/23 03:27, Sean Christopherson wrote:
>>> On Mon, Oct 16, 2023, Dionna Amalie Glaze wrote:
>>>>> +
>>>>> + /*
>>>>> + * If a VMM-specific certificate blob hasn't been provided, grab the
>>>>> + * host-wide one.
>>>>> + */
>>>>> + snp_certs = sev_snp_certs_get(sev->snp_certs);
>>>>> + if (!snp_certs)
>>>>> + snp_certs = sev_snp_global_certs_get();
>>>>> +
>>>>
>>>> This is where the generation I suggested adding would get checked. If
>>>> the instance certs' generation is not the global generation, then I
>>>> think we need a way to return to the VMM to make that right before
>>>> continuing to provide outdated certificates.
>>>> This might be an unreasonable request, but the fact that the certs and
>>>> reported_tcb can be set while a VM is running makes this an issue.
>>>
>>> Before we get that far, the changelogs need to explain why the kernel is storing
>>> userspace blobs in the first place. The whole thing is a bit of a mess.
>>>
>>> sev_snp_global_certs_get() has data races that could lead to variations of TOCTOU
>>> bugs: sev_ioctl_snp_set_config() can overwrite psp_master->sev_data->snp_certs
>>> while sev_snp_global_certs_get() is running. If the compiler reloads snp_certs
>>> between bumping the refcount and grabbing the pointer, KVM will end up leaking a
>>> refcount and consuming a pointer without a refcount.
>>>
>>> if (!kref_get_unless_zero(&certs->kref))
>>> return NULL;
>>>
>>> return certs;
>>
>> I'm missing something here. The @certs pointer is on the stack,
>
> No, nothing guarantees that @certs is on the stack and will never be reloaded.
> sev_snp_certs_get() is in full view of sev_snp_global_certs_get(), so it's entirely
> possible that it can be inlined. Then you end up with:
>
> struct sev_device *sev;
>
> if (!psp_master || !psp_master->sev_data)
> return NULL;
>
> sev = psp_master->sev_data;
> if (!sev->snp_initialized)
> return NULL;
>
> if (!sev->snp_certs)
> return NULL;
>
> if (!kref_get_unless_zero(&sev->snp_certs->kref))
> return NULL;
>
> return sev->snp_certs;
>
> At which point the compiler could choose to omit a local variable entirely, it
> could store @certs in a register and reload after kref_get_unless_zero(), etc.
> If psp_master->sev_data->snp_certs is changed at any point, odd thing can happen.
>
> That atomic operation in kref_get_unless_zero() might prevent a reload between
> getting the kref and the return, but it wouldn't prevent a reload between the
> !NULL check and kref_get_unless_zero().
Oh. The function is exported so I thought gcc would not go that far but
yeah it is possible. So this needs an explicit READ_ONCE barrier.
>>> If userspace wants to provide garbage to the guest, so be it, not KVM's problem.
>>> That way, whether the VM gets the global cert or a per-VM cert is purely a userspace
>>> concern.
>>
>> The global cert lives in CCP (/dev/sev), the per VM cert lives in kvmvm_fd.
>> "A la vcpu->run" is fine for the latter but for the former we need something
>> else.
>
> Why? The cert ultimately comes from userspace, no? Make userspace deal with it.
>
>> And there is scenario when one global certs blob is what is needed and
>> copying it over multiple VMs seems suboptimal.
>
> That's a solvable problem. I'm not sure I like the most obvious solution, but it
> is a solution: let userspace define a KVM-wide blob pointer, either via .mmap()
> or via an ioctl().
>
> FWIW, there's no need to do .mmap() shenanigans, e.g. an ioctl() to set the
> userspace pointer would suffice. The benefit of a kernel controlled pointer is
> that it doesn't require copying to a kernel buffer (or special code to copy from
> userspace into guest).
Just to clarify - like, a small userspace non-qemu program which just
holds a pointer with the certs blob, or embed it into libvirt or systemd?
> Actually, looking at the flow again, AFAICT there's nothing special about the
> target DATA_PAGE. It must be SHARED *before* SVM_VMGEXIT_EXT_GUEST_REQUEST, i.e.
> KVM doesn't need to do conversions, there's no kernel priveleges required, etc.
> And the GHCB doesn't dictate ordering between storing the certificates and doing
> the request. That means the certificate stuff can be punted entirely to usersepace.
All true.
> Heh, typing up the below, there's another bug: KVM will incorrectly "return" '0'
> for non-SNP guests:
>
> unsigned long exitcode = 0;
> u64 data_gpa;
> int err, rc;
>
> if (!sev_snp_guest(vcpu->kvm)) {
> rc = SEV_RET_INVALID_GUEST; <= sets "rc", not "exitcode"
> goto e_fail;
> }
>
> e_fail:
> ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, exitcode);
>
> Which really highlights that we need to get test infrastructure up and running
> for SEV-ES, SNP, and TDX.
>
> Anyways, back to punting to userspace. Here's a rough sketch. The only new uAPI
> is the definition of KVM_HC_SNP_GET_CERTS and its arguments.
>
> static void snp_handle_guest_request(struct vcpu_svm *svm)
> {
> struct vmcb_control_area *control = &svm->vmcb->control;
> struct sev_data_snp_guest_request data = {0};
> struct kvm_vcpu *vcpu = &svm->vcpu;
> struct kvm *kvm = vcpu->kvm;
> struct kvm_sev_info *sev;
> gpa_t req_gpa = control->exit_info_1;
> gpa_t resp_gpa = control->exit_info_2;
> unsigned long rc;
> int err;
>
> if (!sev_snp_guest(vcpu->kvm)) {
> rc = SEV_RET_INVALID_GUEST;
> goto e_fail;
> }
>
> sev = &to_kvm_svm(kvm)->sev_info;
>
> mutex_lock(&sev->guest_req_lock);
>
> rc = snp_setup_guest_buf(svm, &data, req_gpa, resp_gpa);
> if (rc)
> goto unlock;
>
> rc = sev_issue_cmd(kvm, SEV_CMD_SNP_GUEST_REQUEST, &data, &err);
> if (rc)
> /* Ensure an error value is returned to guest. */
> rc = err ? err : SEV_RET_INVALID_ADDRESS;
>
> snp_cleanup_guest_buf(&data, &rc);
>
> unlock:
> mutex_unlock(&sev->guest_req_lock);
>
> e_fail:
> ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, rc);
> }
>
> static int snp_complete_ext_guest_request(struct kvm_vcpu *vcpu)
> {
> u64 certs_exitcode = vcpu->run->hypercall.args[2];
> struct vcpu_svm *svm = to_svm(vcpu);
>
> if (certs_exitcode)
> ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, certs_exitcode);
> else
> snp_handle_guest_request(svm);
> return 1;
> }
>
> static int snp_handle_ext_guest_request(struct vcpu_svm *svm)
> {
> struct kvm_vcpu *vcpu = &svm->vcpu;
> struct kvm *kvm = vcpu->kvm;
> struct kvm_sev_info *sev;
> unsigned long exitcode;
> u64 data_gpa;
>
> if (!sev_snp_guest(vcpu->kvm)) {
> ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, SEV_RET_INVALID_GUEST);
> return 1;
> }
>
> data_gpa = vcpu->arch.regs[VCPU_REGS_RAX];
> if (!IS_ALIGNED(data_gpa, PAGE_SIZE)) {
> ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, SEV_RET_INVALID_ADDRESS);
> return 1;
> }
>
> vcpu->run->hypercall.nr = KVM_HC_SNP_GET_CERTS;
> vcpu->run->hypercall.args[0] = data_gpa;
> vcpu->run->hypercall.args[1] = vcpu->arch.regs[VCPU_REGS_RBX];
> vcpu->run->hypercall.flags = KVM_EXIT_HYPERCALL_LONG_MODE;
btw why is it _LONG_MODE and not just _64? :)
> vcpu->arch.complete_userspace_io = snp_complete_ext_guest_request;
> return 0;
> }
This should work the KVM stored certs nicely but not for the global
certs. Although I am not all convinced that global certs is all that
valuable but I do not know the history of that, happened before I joined
so I let others to comment on that. Thanks,
--
Alexey
next prev parent reply other threads:[~2023-10-19 2:49 UTC|newest]
Thread overview: 158+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-16 13:27 [PATCH v10 00/50] Add AMD Secure Nested Paging (SEV-SNP) Hypervisor Support Michael Roth
2023-10-16 13:27 ` [PATCH v10 01/50] KVM: SVM: INTERCEPT_RDTSCP is never intercepted anyway Michael Roth
2023-10-16 15:12 ` Greg KH
2023-10-16 15:14 ` Paolo Bonzini
2023-10-16 15:21 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 02/50] KVM: SVM: Fix TSC_AUX virtualization setup Michael Roth
2023-10-16 13:27 ` [PATCH v10 03/50] KVM: SEV: Do not intercept accesses to MSR_IA32_XSS for SEV-ES guests Michael Roth
2023-12-13 12:50 ` Paolo Bonzini
2023-12-13 17:30 ` Sean Christopherson
2023-12-13 17:40 ` Paolo Bonzini
2023-10-16 13:27 ` [PATCH v10 04/50] x86/cpufeatures: Add SEV-SNP CPU feature Michael Roth
2023-12-13 12:51 ` Paolo Bonzini
2023-12-13 13:13 ` Borislav Petkov
2023-12-13 13:31 ` Paolo Bonzini
2023-12-13 13:36 ` Borislav Petkov
2023-12-13 13:40 ` Paolo Bonzini
2023-12-13 13:49 ` Borislav Petkov
2023-12-13 14:18 ` Paolo Bonzini
2023-12-13 15:41 ` Borislav Petkov
2023-12-13 17:35 ` Paolo Bonzini
2023-12-13 18:53 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 05/50] x86/speculation: Do not enable Automatic IBRS if SEV SNP is enabled Michael Roth
2023-10-25 17:33 ` Borislav Petkov
2023-10-27 21:50 ` Dave Hansen
2023-12-13 12:52 ` Paolo Bonzini
2023-10-16 13:27 ` [PATCH v10 06/50] x86/sev: Add the host SEV-SNP initialization support Michael Roth
2023-10-25 18:19 ` Tom Lendacky
2023-11-07 16:31 ` Borislav Petkov
2023-11-07 18:32 ` Tom Lendacky
2023-11-07 19:13 ` Borislav Petkov
2023-11-08 8:21 ` Jeremi Piotrowski
2023-11-08 15:19 ` Tom Lendacky
2023-11-07 19:00 ` Kalra, Ashish
2023-11-07 19:19 ` Borislav Petkov
2023-11-07 20:27 ` Borislav Petkov
2023-11-07 21:21 ` Kalra, Ashish
2023-11-07 21:27 ` Borislav Petkov
2023-11-07 22:08 ` Borislav Petkov
2023-11-07 22:33 ` Kalra, Ashish
2023-11-08 6:14 ` Borislav Petkov
2023-11-08 9:11 ` Jeremi Piotrowski
2023-11-08 19:53 ` Kalra, Ashish
2023-12-08 17:09 ` Jeremi Piotrowski
2023-12-08 23:21 ` Kalra, Ashish
2023-12-20 7:07 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 07/50] x86/sev: Add RMP entry lookup helpers Michael Roth
2023-11-14 14:24 ` Borislav Petkov
2023-12-19 3:31 ` Michael Roth
2024-01-09 22:07 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 08/50] x86/fault: Add helper for dumping RMP entries Michael Roth
2023-11-15 16:08 ` Borislav Petkov
2023-12-19 6:08 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 09/50] x86/traps: Define RMP violation #PF error code Michael Roth
2023-10-16 14:14 ` Dave Hansen
2023-10-16 14:55 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 10/50] x86/fault: Report RMP page faults for kernel addresses Michael Roth
2023-11-21 15:23 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 11/50] x86/sev: Add helper functions for RMPUPDATE and PSMASH instruction Michael Roth
2023-11-21 16:21 ` Borislav Petkov
2023-12-19 16:20 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 12/50] x86/sev: Invalidate pages from the direct map when adding them to the RMP table Michael Roth
2023-11-24 14:20 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 13/50] crypto: ccp: Define the SEV-SNP commands Michael Roth
2023-11-24 14:36 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 14/50] crypto: ccp: Add support to initialize the AMD-SP for SEV-SNP Michael Roth
2023-11-27 9:59 ` Borislav Petkov
2023-11-30 2:13 ` Kalra, Ashish
2023-12-06 17:08 ` Borislav Petkov
2023-12-06 20:35 ` Kalra, Ashish
2023-12-09 16:20 ` Borislav Petkov
2023-12-11 21:11 ` Kalra, Ashish
2023-12-12 6:52 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 15/50] crypto: ccp: Provide API to issue SEV and SNP commands Michael Roth
2023-12-06 20:21 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 16/50] x86/sev: Introduce snp leaked pages list Michael Roth
2023-12-06 20:42 ` Borislav Petkov
2023-12-08 20:54 ` Kalra, Ashish
2023-12-07 16:20 ` Vlastimil Babka
2023-12-08 22:10 ` Kalra, Ashish
2023-12-11 13:08 ` Vlastimil Babka
2023-12-12 23:26 ` Kalra, Ashish
2023-10-16 13:27 ` [PATCH v10 17/50] crypto: ccp: Handle the legacy TMR allocation when SNP is enabled Michael Roth
2023-12-08 13:05 ` Borislav Petkov
2023-12-19 23:46 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 18/50] crypto: ccp: Handle the legacy SEV command " Michael Roth
2023-12-09 15:36 ` Borislav Petkov
2023-12-29 21:38 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 19/50] crypto: ccp: Add the SNP_PLATFORM_STATUS command Michael Roth
2023-12-12 16:45 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 20/50] KVM: SEV: Select CONFIG_KVM_SW_PROTECTED_VM when CONFIG_KVM_AMD_SEV=y Michael Roth
2023-12-13 12:54 ` Paolo Bonzini
2023-12-29 21:41 ` Michael Roth
2023-12-18 10:13 ` Borislav Petkov
2023-12-29 21:40 ` Michael Roth
2023-10-16 13:27 ` [PATCH v10 21/50] KVM: SEV: Add support to handle AP reset MSR protocol Michael Roth
2023-12-12 17:02 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 22/50] KVM: SEV: Add GHCB handling for Hypervisor Feature Support requests Michael Roth
2023-12-18 10:23 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 23/50] KVM: SEV: Make AVIC backing, VMSA and VMCB memory allocation SNP safe Michael Roth
2023-12-11 13:24 ` Vlastimil Babka
2023-12-12 0:00 ` Kalra, Ashish
2023-12-13 13:31 ` Paolo Bonzini
2023-12-13 18:45 ` Paolo Bonzini
2023-12-18 14:57 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 24/50] KVM: SEV: Add initial SEV-SNP support Michael Roth
2023-12-18 17:43 ` Borislav Petkov
2023-10-16 13:27 ` [PATCH v10 25/50] KVM: SEV: Add KVM_SNP_INIT command Michael Roth
2023-10-16 13:27 ` [PATCH v10 26/50] KVM: SEV: Add KVM_SEV_SNP_LAUNCH_START command Michael Roth
2023-10-16 13:27 ` [PATCH v10 27/50] KVM: Add HVA range operator Michael Roth
2023-10-16 13:27 ` [PATCH v10 28/50] KVM: SEV: Add KVM_SEV_SNP_LAUNCH_UPDATE command Michael Roth
2023-10-16 13:27 ` [PATCH v10 29/50] KVM: SEV: Add KVM_SEV_SNP_LAUNCH_FINISH command Michael Roth
2023-10-16 13:27 ` [PATCH v10 30/50] KVM: SEV: Add support to handle GHCB GPA register VMGEXIT Michael Roth
2023-10-16 13:28 ` [PATCH v10 31/50] KVM: SEV: Add KVM_EXIT_VMGEXIT Michael Roth
2023-10-16 13:28 ` [PATCH v10 32/50] KVM: SEV: Add support to handle MSR based Page State Change VMGEXIT Michael Roth
2023-10-16 13:28 ` [PATCH v10 33/50] KVM: SEV: Add support to handle " Michael Roth
2023-10-16 13:28 ` [PATCH v10 34/50] KVM: x86: Export the kvm_zap_gfn_range() for the SNP use Michael Roth
2023-10-16 13:28 ` [PATCH v10 35/50] KVM: SEV: Add support to handle RMP nested page faults Michael Roth
2023-10-16 13:28 ` [PATCH v10 36/50] KVM: SEV: Use a VMSA physical address variable for populating VMCB Michael Roth
2023-10-16 13:28 ` [PATCH v10 37/50] KVM: SEV: Support SEV-SNP AP Creation NAE event Michael Roth
2023-10-16 13:28 ` [PATCH v10 38/50] KVM: SEV: Add support for GHCB-based termination requests Michael Roth
2023-10-19 12:20 ` Liam Merwick
2023-10-16 13:28 ` [PATCH v10 39/50] KVM: SEV: Implement gmem hook for initializing private pages Michael Roth
2023-10-16 13:28 ` [PATCH v10 40/50] KVM: SEV: Implement gmem hook for invalidating " Michael Roth
2023-10-16 13:28 ` [PATCH v10 41/50] KVM: x86: Add gmem hook for determining max NPT mapping level Michael Roth
2023-10-16 13:28 ` [PATCH v10 42/50] KVM: SEV: Avoid WBINVD for HVA-based MMU notifications for SNP Michael Roth
2023-10-16 13:28 ` [PATCH v10 43/50] KVM: SVM: Add module parameter to enable the SEV-SNP Michael Roth
2023-10-16 13:28 ` [PATCH v10 44/50] iommu/amd: Add IOMMU_SNP_SHUTDOWN support Michael Roth
2023-10-16 13:28 ` [PATCH v10 45/50] iommu/amd: Report all cases inhibiting SNP enablement Michael Roth
2023-10-16 13:28 ` [PATCH v10 46/50] crypto: ccp: Add the SNP_{SET,GET}_EXT_CONFIG command Michael Roth
2023-10-16 23:11 ` Dionna Amalie Glaze
2023-10-16 13:28 ` [PATCH v10 47/50] x86/sev: Add KVM commands for per-instance certs Michael Roth
2023-10-16 13:28 ` [PATCH v10 48/50] KVM: SEV: Provide support for SNP_GUEST_REQUEST NAE event Michael Roth
2023-10-16 23:18 ` Dionna Amalie Glaze
2023-10-17 16:27 ` Sean Christopherson
2023-10-18 2:28 ` Alexey Kardashevskiy
2023-10-18 13:48 ` Sean Christopherson
2023-10-18 20:27 ` Kalra, Ashish
2023-10-18 20:38 ` Sean Christopherson
2023-10-18 21:27 ` Kalra, Ashish
2023-10-18 21:43 ` Sean Christopherson
2023-10-19 2:48 ` Alexey Kardashevskiy [this message]
2023-10-19 14:57 ` Sean Christopherson
2023-10-19 23:55 ` Alexey Kardashevskiy
2023-10-20 0:13 ` Sean Christopherson
2023-10-20 0:43 ` Alexey Kardashevskiy
2023-10-20 15:13 ` Sean Christopherson
2023-10-20 18:37 ` Tom Lendacky
2023-11-10 22:07 ` Michael Roth
2023-11-10 22:47 ` Sean Christopherson
2023-11-16 5:31 ` Dionna Amalie Glaze
2023-12-05 0:30 ` Dan Williams
2023-12-05 0:48 ` Dionna Amalie Glaze
2023-12-05 20:06 ` Dan Williams
2023-12-05 22:04 ` Dionna Amalie Glaze
2023-12-05 23:11 ` Dan Williams
2023-12-06 0:43 ` Dionna Amalie Glaze
2023-10-16 13:28 ` [PATCH v10 49/50] crypto: ccp: Add debug support for decrypting pages Michael Roth
2023-10-16 13:28 ` [PATCH v10 50/50] crypto: ccp: Add panic notifier for SEV/SNP firmware shutdown on kdump Michael Roth
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=924b755a-977a-4476-9525-a7626d728e18@amd.com \
--to=aik@amd.com \
--cc=ak@linux.intel.com \
--cc=alpergun@google.com \
--cc=ardb@kernel.org \
--cc=ashish.kalra@amd.com \
--cc=bp@alien8.de \
--cc=brijesh.singh@amd.com \
--cc=dave.hansen@linux.intel.com \
--cc=dionnaglaze@google.com \
--cc=dovmurik@linux.ibm.com \
--cc=hpa@zytor.com \
--cc=jarkko@kernel.org \
--cc=jmattson@google.com \
--cc=jroedel@suse.de \
--cc=kirill@shutemov.name \
--cc=kvm@vger.kernel.org \
--cc=liam.merwick@oracle.com \
--cc=linux-coco@lists.linux.dev \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=marcorr@google.com \
--cc=michael.roth@amd.com \
--cc=mingo@redhat.com \
--cc=nikunj.dadhania@amd.com \
--cc=pankaj.gupta@amd.com \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=pgonda@google.com \
--cc=rientjes@google.com \
--cc=sathyanarayanan.kuppuswamy@linux.intel.com \
--cc=seanjc@google.com \
--cc=slp@redhat.com \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=tglx@linutronix.de \
--cc=thomas.lendacky@amd.com \
--cc=tobin@ibm.com \
--cc=tony.luck@intel.com \
--cc=vbabka@suse.cz \
--cc=vkuznets@redhat.com \
--cc=x86@kernel.org \
--cc=zhi.a.wang@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).