All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 0/9] Add AMD SEV page encryption bitmap support.
@ 2020-12-01  0:45 Ashish Kalra
  2020-12-01  0:45 ` [PATCH v2 1/9] KVM: x86: Add AMD SEV specific Hypercall3 Ashish Kalra
                   ` (8 more replies)
  0 siblings, 9 replies; 39+ messages in thread
From: Ashish Kalra @ 2020-12-01  0:45 UTC (permalink / raw)
  To: pbonzini
  Cc: tglx, mingo, hpa, joro, bp, thomas.lendacky, x86, kvm,
	linux-kernel, srutherford, brijesh.singh, dovmurik, tobin, jejb,
	frankeh, dgilbert

From: Ashish Kalra <ashish.kalra@amd.com>

The series add support for AMD SEV page encryption bitmap.

SEV guest VMs have the concept of private and shared memory. Private memory
is encrypted with the guest-specific key, while shared memory may be encrypted
with hypervisor key. The patch series introduces a new hypercall.
The guest OS can use this hypercall to notify the page encryption status.

The patch adds new ioctls KVM_{SET,GET}_PAGE_ENC_BITMAP. The ioctl can be used
by qemu to get the page encryption bitmap. Qemu can consult this bitmap
during guest live migration / page migration and/or guest debugging to know
whether the page is encrypted.

The page encryption bitmap support is required for SEV guest live migration,
guest page migration and guest debugging.

The patch-set also adds support for bypassing unencrypted guest memory
regions for DBG_DECRYPT API calls, guest memory region encryption status
in sev_dbg_decrypt() is now referenced using the page encryption bitmap.

A branch containing these patches is available here:
https://github.com/AMDESE/linux/tree/sev-page-encryption-bitmap-v2

Changes since v1:
 - Fix in sev_dbg_crypt() to release RCU read lock if hva_to_gfn() fails
   when bypassing DBG_DECRYPT API calls for unencrypted guest memory.
 - Comment fix for Patch 7/9. 

Ashish Kalra (4):
  KVM: SVM: Add support for static allocation of unified Page Encryption
    Bitmap.
  KVM: x86: Mark _bss_decrypted section variables as decrypted in page
    encryption bitmap.
  KVM: x86: Add kexec support for SEV page encryption bitmap.
  KVM: SVM: Bypass DBG_DECRYPT API calls for unecrypted guest memory.

Brijesh Singh (5):
  KVM: x86: Add AMD SEV specific Hypercall3
  KVM: X86: Introduce KVM_HC_PAGE_ENC_STATUS hypercall
  KVM: x86: Introduce KVM_GET_PAGE_ENC_BITMAP ioctl
  mm: x86: Invoke hypercall when page encryption status is changed.
  KVM: x86: Introduce KVM_SET_PAGE_ENC_BITMAP ioctl

 Documentation/virt/kvm/api.rst        |  71 ++++++
 Documentation/virt/kvm/hypercalls.rst |  15 ++
 arch/x86/include/asm/kvm_host.h       |   7 +
 arch/x86/include/asm/kvm_para.h       |  12 +
 arch/x86/include/asm/mem_encrypt.h    |   4 +
 arch/x86/include/asm/paravirt.h       |  10 +
 arch/x86/include/asm/paravirt_types.h |   2 +
 arch/x86/kernel/kvm.c                 |  28 +++
 arch/x86/kernel/kvmclock.c            |  12 +
 arch/x86/kernel/paravirt.c            |   1 +
 arch/x86/kvm/svm/sev.c                | 321 ++++++++++++++++++++++++++
 arch/x86/kvm/svm/svm.c                |   5 +
 arch/x86/kvm/svm/svm.h                |   7 +
 arch/x86/kvm/vmx/vmx.c                |   1 +
 arch/x86/kvm/x86.c                    |  35 +++
 arch/x86/mm/mem_encrypt.c             |  63 ++++-
 arch/x86/mm/pat/set_memory.c          |   7 +
 include/uapi/linux/kvm.h              |  13 ++
 include/uapi/linux/kvm_para.h         |   1 +
 19 files changed, 614 insertions(+), 1 deletion(-)

-- 
2.17.1


^ permalink raw reply	[flat|nested] 39+ messages in thread
* Re: [PATCH v2 1/9] KVM: x86: Add AMD SEV specific Hypercall3
@ 2020-12-08  5:18 Kalra, Ashish
  0 siblings, 0 replies; 39+ messages in thread
From: Kalra, Ashish @ 2020-12-08  5:18 UTC (permalink / raw)
  To: Steve Rutherford
  Cc: Sean Christopherson, Paolo Bonzini, Thomas Gleixner, Ingo Molnar,
	H. Peter Anvin, Joerg Roedel, Borislav Petkov, Lendacky, Thomas,
	X86 ML, KVM list, LKML, Singh, Brijesh, dovmurik, tobin, jejb,
	frankeh, dgilbert


> 
>> I suspect a list
>> would consume far less memory, hopefully without impacting performance.

And how much host memory are we talking about for here, say for a 4gb guest, the bitmap will be using just using something like 128k+.

Thanks,
Ashish

> On Dec 7, 2020, at 10:16 PM, Kalra, Ashish <Ashish.Kalra@amd.com> wrote:
> 
> I don’t think that the bitmap by itself is really a performance bottleneck here.
> 
> Thanks,
> Ashish
> 
>>> On Dec 7, 2020, at 9:10 PM, Steve Rutherford <srutherford@google.com> wrote:
>>> On Mon, Dec 7, 2020 at 12:42 PM Sean Christopherson <seanjc@google.com> wrote:
>>>> On Sun, Dec 06, 2020, Paolo Bonzini wrote:
>>>> On 03/12/20 01:34, Sean Christopherson wrote:
>>>>> On Tue, Dec 01, 2020, Ashish Kalra wrote:
>>>>>> From: Brijesh Singh <brijesh.singh@amd.com>
>>>>>> KVM hypercall framework relies on alternative framework to patch the
>>>>>> VMCALL -> VMMCALL on AMD platform. If a hypercall is made before
>>>>>> apply_alternative() is called then it defaults to VMCALL. The approach
>>>>>> works fine on non SEV guest. A VMCALL would causes #UD, and hypervisor
>>>>>> will be able to decode the instruction and do the right things. But
>>>>>> when SEV is active, guest memory is encrypted with guest key and
>>>>>> hypervisor will not be able to decode the instruction bytes.
>>>>>> Add SEV specific hypercall3, it unconditionally uses VMMCALL. The hypercall
>>>>>> will be used by the SEV guest to notify encrypted pages to the hypervisor.
>>>>> What if we invert KVM_HYPERCALL and X86_FEATURE_VMMCALL to default to VMMCALL
>>>>> and opt into VMCALL?  It's a synthetic feature flag either way, and I don't
>>>>> think there are any existing KVM hypercalls that happen before alternatives are
>>>>> patched, i.e. it'll be a nop for sane kernel builds.
>>>>> I'm also skeptical that a KVM specific hypercall is the right approach for the
>>>>> encryption behavior, but I'll take that up in the patches later in the series.
>>>> Do you think that it's the guest that should "donate" memory for the bitmap
>>>> instead?
>>> No.  Two things I'd like to explore:
>>> 1. Making the hypercall to announce/request private vs. shared common across
>>>   hypervisors (KVM, Hyper-V, VMware, etc...) and technologies (SEV-* and TDX).
>>>   I'm concerned that we'll end up with multiple hypercalls that do more or
>>>   less the same thing, e.g. KVM+SEV, Hyper-V+SEV, TDX, etc...  Maybe it's a
>>>   pipe dream, but I'd like to at least explore options before shoving in KVM-
>>>   only hypercalls.
>>> 2. Tracking shared memory via a list of ranges instead of a using bitmap to
>>>   track all of guest memory.  For most use cases, the vast majority of guest
>>>   memory will be private, most ranges will be 2mb+, and conversions between
>>>   private and shared will be uncommon events, i.e. the overhead to walk and
>>>   split/merge list entries is hopefully not a big concern.  I suspect a list
>>>   would consume far less memory, hopefully without impacting performance.
>> For a fancier data structure, I'd suggest an interval tree. Linux
>> already has an rbtree-based interval tree implementation, which would
>> likely work, and would probably assuage any performance concerns.
>> Something like this would not be worth doing unless most of the shared
>> pages were physically contiguous. A sample Ubuntu 20.04 VM on GCP had
>> 60ish discontiguous shared regions. This is by no means a thorough
>> search, but it's suggestive. If this is typical, then the bitmap would
>> be far less efficient than most any interval-based data structure.
>> You'd have to allow userspace to upper bound the number of intervals
>> (similar to the maximum bitmap size), to prevent host OOMs due to
>> malicious guests. There's something nice about the guest donating
>> memory for this, since that would eliminate the OOM risk.

^ permalink raw reply	[flat|nested] 39+ messages in thread
end of thread, other threads:[~2021-01-08 16:57 UTC | newest]

Thread overview: 39+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-12-01  0:45 [PATCH v2 0/9] Add AMD SEV page encryption bitmap support Ashish Kalra
2020-12-01  0:45 ` [PATCH v2 1/9] KVM: x86: Add AMD SEV specific Hypercall3 Ashish Kalra
2020-12-03  0:34   ` Sean Christopherson
2020-12-04 17:16     ` Brijesh Singh
2020-12-06 10:26     ` Paolo Bonzini
2020-12-07 20:41       ` Sean Christopherson
2020-12-08  3:09         ` Steve Rutherford
2020-12-08  4:16           ` Kalra, Ashish
2020-12-08 16:29           ` Brijesh Singh
2020-12-11 22:55             ` Ashish Kalra
2020-12-12  4:56               ` Ashish Kalra
2020-12-18 19:39                 ` Dr. David Alan Gilbert
     [not found]                   ` <E79E09A2-F314-4B59-B7AE-07B1D422DF2B@amd.com>
2020-12-18 19:56                     ` Dr. David Alan Gilbert
2021-01-06 23:05                       ` Ashish Kalra
2021-01-07  1:01                         ` Steve Rutherford
2021-01-07  1:34                           ` Ashish Kalra
2021-01-07  8:05                             ` Ashish Kalra
2021-01-08  0:47                               ` Ashish Kalra
2021-01-08  0:55                                 ` Steve Rutherford
2021-01-07 17:07                           ` Ashish Kalra
2021-01-07 17:26                             ` Sean Christopherson
2021-01-07 18:41                               ` Ashish Kalra
2021-01-07 19:22                                 ` Sean Christopherson
2021-01-08  0:54                                   ` Steve Rutherford
2021-01-08 16:56                                     ` Sean Christopherson
2020-12-01  0:46 ` [PATCH v2 2/9] KVM: X86: Introduce KVM_HC_PAGE_ENC_STATUS hypercall Ashish Kalra
2020-12-02 16:54   ` Dr. David Alan Gilbert
2020-12-02 21:22     ` Ashish Kalra
2020-12-06 10:25       ` Paolo Bonzini
2020-12-01  0:47 ` [PATCH v2 3/9] KVM: x86: Introduce KVM_GET_PAGE_ENC_BITMAP ioctl Ashish Kalra
2020-12-06 11:02   ` Dov Murik
2020-12-07 22:00     ` Ashish Kalra
2020-12-01  0:47 ` [PATCH v2 4/9] mm: x86: Invoke hypercall when page encryption status is changed Ashish Kalra
2020-12-01  0:47 ` [PATCH v2 5/9] KVM: x86: Introduce KVM_SET_PAGE_ENC_BITMAP ioctl Ashish Kalra
2020-12-01  0:47 ` [PATCH v2 6/9] KVM: SVM: Add support for static allocation of unified Page Encryption Bitmap Ashish Kalra
2020-12-01  0:48 ` [PATCH v2 7/9] KVM: x86: Mark _bss_decrypted section variables as decrypted in page encryption bitmap Ashish Kalra
2020-12-01  0:48 ` [PATCH v2 8/9] KVM: x86: Add kexec support for SEV " Ashish Kalra
2020-12-01  0:48 ` [PATCH v2 9/9] KVM: SVM: Bypass DBG_DECRYPT API calls for unecrypted guest memory Ashish Kalra
2020-12-08  5:18 [PATCH v2 1/9] KVM: x86: Add AMD SEV specific Hypercall3 Kalra, Ashish

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.