From: "Michael S. Tsirkin" <mst@redhat.com>
To: Andi Kleen <ak@linux.intel.com>
Cc: linux-kernel@vger.kernel.org,
Kuppuswamy Sathyanarayanan
<sathyanarayanan.kuppuswamy@linux.intel.com>
Subject: Re: [PATCH v5 16/16] x86/tdx: Add cmdline option to force use of ioremap_host_shared
Date: Sun, 17 Oct 2021 18:34:56 -0400 [thread overview]
Message-ID: <20211017182145-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <d5e755fd-41a7-7c3f-f540-094c1c583365@linux.intel.com>
On Fri, Oct 15, 2021 at 06:34:17AM -0700, Andi Kleen wrote:
> cutting down the insane cc list.
>
> On 10/14/2021 11:57 PM, Michael S. Tsirkin wrote:
> > On Thu, Oct 14, 2021 at 10:50:59PM -0700, Andi Kleen wrote:
> > > > I thought you basically create an OperationRegion of SystemMemory type,
> > > > and off you go. Maybe the OSPM in Linux is clever and protects
> > > > some memory, I wouldn't know.
> > >
> > > I investigated this now, and it looks like acpi is using ioremap_cache(). We
> > > can hook into that and force non sharing. It's probably safe to assume that
> > > this is not used on real IO devices.
> > >
> > > I think there are still some other BIOS mappings that use just plain
> > > ioremap() though.
> > >
> > >
> > > -Andi
> > Hmm don't you mean the reverse? If you make ioremap shared then OS is
> > protected from malicious ACPI?
>
>
> Nope
>
> > If you don't make it shared then
> > malicious ACPI can poke at arbitrary OS memory.
>
>
> When it's private it's protected and when it's shared it can be attacked
>
>
> >
> > For BIOS I suspect there's no way around it, it needs to be
> > audited since it's executable.
>
>
> The guest BIOS is attested and trusted. The original ACPI tables by the BIOS
> are attested and trusted too.
>
> Just if we map the ACPI tables temporarily shared then an evil hypervisor
> could modify them during that time window.
>
> -Andi
I thought some more about it.
Fundamentally, ACPI has these types of OperationRegions:
SystemIO | SystemMemory | PCI_Config | EmbeddedControl | SMBus | SystemCMOS |
PciBarTarget | IPMI | GeneralPurposeIO | GenericSerialBus |
PCC
Now, SystemMemory can be used to talk to either BIOS (should be
encrypted) or hypervisor (should not be encrypted).
I think it's not a great idea to commit to either, or teach users
to hack around it with command line flags. Instead
there should be a new SystemMemoryUnencrypted API for interface with
the hypervisor. Can you guys propose this at the ACPI spec?
If not but at least we are in agreement I guess I can try to do it,
have a bit of experience with the ACPI spec.
And I assume PciBarTarget should be unencrypted so it can work.
--
MST
prev parent reply other threads:[~2021-10-17 22:41 UTC|newest]
Thread overview: 85+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-09 0:36 [PATCH v5 00/16] Add TDX Guest Support (shared-mm support) Kuppuswamy Sathyanarayanan
2021-10-09 0:36 ` [PATCH v5 01/16] x86/mm: Move force_dma_unencrypted() to common code Kuppuswamy Sathyanarayanan
2021-10-20 16:11 ` Tom Lendacky
2021-10-20 16:43 ` Sathyanarayanan Kuppuswamy
2021-10-09 0:36 ` [PATCH v5 02/16] x86/tdx: Get TD execution environment information via TDINFO Kuppuswamy Sathyanarayanan
2021-10-09 0:36 ` [PATCH v5 03/16] x86/tdx: Exclude Shared bit from physical_mask Kuppuswamy Sathyanarayanan
2021-11-05 22:11 ` Sean Christopherson
2021-11-08 14:45 ` Kirill A. Shutemov
2021-10-09 0:36 ` [PATCH v5 04/16] x86/tdx: Make pages shared in ioremap() Kuppuswamy Sathyanarayanan
2021-10-20 16:03 ` Tom Lendacky
2021-10-20 16:41 ` Sathyanarayanan Kuppuswamy
2021-10-09 0:37 ` [PATCH v5 05/16] x86/tdx: Add helper to do MapGPA hypercall Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 06/16] x86/tdx: Make DMA pages shared Kuppuswamy Sathyanarayanan
2021-10-20 16:33 ` Tom Lendacky
2021-10-20 16:45 ` Sathyanarayanan Kuppuswamy
2021-10-20 17:22 ` Tom Lendacky
2021-10-20 17:26 ` Sathyanarayanan Kuppuswamy
2021-10-09 0:37 ` [PATCH v5 07/16] x86/kvm: Use bounce buffers for TD guest Kuppuswamy Sathyanarayanan
2021-10-20 16:39 ` Tom Lendacky
2021-10-20 16:50 ` Sathyanarayanan Kuppuswamy
2021-10-20 17:26 ` Tom Lendacky
2021-10-09 0:37 ` [PATCH v5 08/16] x86/tdx: ioapic: Add shared bit for IOAPIC base address Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 09/16] x86/tdx: Enable shared memory confidential guest flags for TDX guest Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 10/16] PCI: Consolidate pci_iomap_range(), pci_iomap_wc_range() Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 11/16] asm/io.h: Add ioremap_host_shared fallback Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 12/16] PCI: Add pci_iomap_host_shared(), pci_iomap_host_shared_range() Kuppuswamy Sathyanarayanan
2021-10-09 9:53 ` Michael S. Tsirkin
2021-10-09 20:39 ` Dan Williams
2021-10-10 22:11 ` Andi Kleen
2021-10-12 17:42 ` Dan Williams
2021-10-12 18:35 ` Andi Kleen
2021-10-12 21:14 ` Dan Williams
2021-10-12 21:18 ` Michael S. Tsirkin
2021-10-12 21:24 ` Andi Kleen
2021-10-12 21:28 ` Andi Kleen
2021-10-12 22:00 ` Dan Williams
2021-10-18 12:13 ` Greg KH
2021-10-12 18:36 ` Reshetova, Elena
2021-10-12 18:38 ` Andi Kleen
2021-10-12 18:57 ` Reshetova, Elena
2021-10-12 19:13 ` Dan Williams
2021-10-12 19:49 ` Andi Kleen
2021-10-12 21:11 ` Michael S. Tsirkin
2021-10-14 6:32 ` Reshetova, Elena
2021-10-14 6:57 ` Michael S. Tsirkin
2021-10-14 7:27 ` Reshetova, Elena
2021-10-14 9:26 ` Michael S. Tsirkin
2021-10-14 12:33 ` Reshetova, Elena
2021-10-17 22:17 ` Michael S. Tsirkin
2021-10-14 11:49 ` Michael S. Tsirkin
2021-10-17 21:52 ` Thomas Gleixner
2021-10-18 7:03 ` Reshetova, Elena
2021-10-18 0:55 ` Thomas Gleixner
2021-10-18 1:10 ` Thomas Gleixner
2021-10-18 12:08 ` Greg KH
2021-10-10 22:22 ` Andi Kleen
2021-10-11 11:59 ` Michael S. Tsirkin
2021-10-11 17:32 ` Andi Kleen
2021-10-11 18:22 ` Michael S. Tsirkin
2021-10-18 12:15 ` Greg KH
2021-10-18 13:17 ` Michael S. Tsirkin
2021-10-11 7:58 ` Christoph Hellwig
2021-10-11 17:23 ` Andi Kleen
2021-10-11 19:09 ` Michael S. Tsirkin
2021-10-12 5:31 ` Christoph Hellwig
2021-10-12 18:37 ` Andi Kleen
2021-10-09 0:37 ` [PATCH v5 13/16] PCI: Mark MSI data shared Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 14/16] virtio: Use shared mappings for virtio PCI devices Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 15/16] x86/tdx: Implement ioremap_host_shared for x86 Kuppuswamy Sathyanarayanan
2021-10-09 0:37 ` [PATCH v5 16/16] x86/tdx: Add cmdline option to force use of ioremap_host_shared Kuppuswamy Sathyanarayanan
2021-10-09 1:45 ` Randy Dunlap
2021-10-09 2:10 ` Kuppuswamy, Sathyanarayanan
2021-10-09 11:04 ` Michael S. Tsirkin
2021-10-11 2:39 ` Andi Kleen
2021-10-11 12:04 ` Michael S. Tsirkin
2021-10-11 17:35 ` Andi Kleen
2021-10-11 18:28 ` Michael S. Tsirkin
2021-10-12 17:55 ` Andi Kleen
2021-10-12 20:59 ` Michael S. Tsirkin
2021-10-12 21:18 ` Andi Kleen
2021-10-12 21:30 ` Michael S. Tsirkin
2021-10-15 5:50 ` Andi Kleen
2021-10-15 6:57 ` Michael S. Tsirkin
2021-10-15 13:34 ` Andi Kleen
2021-10-17 22:34 ` Michael S. Tsirkin [this message]
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=20211017182145-mutt-send-email-mst@kernel.org \
--to=mst@redhat.com \
--cc=ak@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sathyanarayanan.kuppuswamy@linux.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).