From: Tom Lendacky <thomas.lendacky@amd.com>
To: Halil Pasic <pasic@linux.ibm.com>, "Michael S. Tsirkin" <mst@redhat.com>
Cc: linux-s390@vger.kernel.org, "Huang, Wei" <wei.huang2@amd.com>,
"Singh, Brijesh" <brijesh.singh@amd.com>,
Janosch Frank <frankja@linux.ibm.com>,
Jason Wang <jasowang@redhat.com>,
Cornelia Huck <cohuck@redhat.com>, Ram Pai <linuxram@us.ibm.com>,
linux-kernel@vger.kernel.org,
virtualization@lists.linux-foundation.org,
Christian Borntraeger <borntraeger@de.ibm.com>,
iommu@lists.linux-foundation.org,
David Gibson <david@gibson.dropbear.id.au>,
Michael Mueller <mimu@linux.ibm.com>,
Viktor Mihajlovski <mihajlov@linux.ibm.com>,
Robin Murphy <robin.murphy@arm.com>,
Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 2/2] virtio: let virtio use DMA API when guest RAM is protected
Date: Fri, 21 Feb 2020 09:39:38 -0600 [thread overview]
Message-ID: <bf0c87a2-adc6-a05a-835f-f98e7cd0cd3b@amd.com> (raw)
In-Reply-To: <20200221141230.13eebc35.pasic@linux.ibm.com>
On 2/21/20 7:12 AM, Halil Pasic wrote:
> On Thu, 20 Feb 2020 15:55:14 -0500
> "Michael S. Tsirkin" <mst@redhat.com> wrote:
>
>> On Thu, Feb 20, 2020 at 05:06:06PM +0100, Halil Pasic wrote:
>>> Currently the advanced guest memory protection technologies (AMD SEV,
>>> powerpc secure guest technology and s390 Protected VMs) abuse the
>>> VIRTIO_F_IOMMU_PLATFORM flag to make virtio core use the DMA API, which
>>> is in turn necessary, to make IO work with guest memory protection.
>>>
>>> But VIRTIO_F_IOMMU_PLATFORM a.k.a. VIRTIO_F_ACCESS_PLATFORM is really a
>>> different beast: with virtio devices whose implementation runs on an SMP
>>> CPU we are still fine with doing all the usual optimizations, it is just
>>> that we need to make sure that the memory protection mechanism does not
>>> get in the way. The VIRTIO_F_ACCESS_PLATFORM mandates more work on the
>>> side of the guest (and possibly he host side as well) than we actually
>>> need.
>>>
>>> An additional benefit of teaching the guest to make the right decision
>>> (and use DMA API) on it's own is: removing the need, to mandate special
>>> VM configuration for guests that may run with protection. This is
>>> especially interesting for s390 as VIRTIO_F_IOMMU_PLATFORM pushes all
>>> the virtio control structures into the first 2G of guest memory:
>>> something we don't necessarily want to do per-default.
>>>
>>> Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
>>> Tested-by: Ram Pai <linuxram@us.ibm.com>
>>> Tested-by: Michael Mueller <mimu@linux.ibm.com>
>>
>> This might work for you but it's fragile, since without
>> VIRTIO_F_ACCESS_PLATFORM hypervisor assumes it gets
>> GPA's, not DMA addresses.
>>
>
> Thanks for your constructive approach. I do want the hypervisor to
> assume it gets GPA's. My train of thought was that the guys that need
> to use IOVA's that are not GPA's when force_dma_unencrypted() will have
> to to specify VIRTIO_F_ACCESS_PLATFORM (at the device) anyway, because
> otherwise it won't work. But I see your point: in case of a
> mis-configuration and provided the DMA API returns IOVA's one could end
> up trying to touch wrong memory locations. But this should be similar to
> what would happen if DMA ops are not used, and memory is not made accessible.
>
>>
>>
>> IOW this looks like another iteration of:
>>
>> virtio: Support encrypted memory on powerpc secure guests
>>
>> which I was under the impression was abandoned as unnecessary.
>
> Unnecessary for powerpc because they do normal PCI. In the context of
> CCW there are only guest physical addresses (CCW I/O has no concept of
> IOMMU or IOVAs).
>
>>
>>
>> To summarize, the necessary conditions for a hack along these lines
>> (using DMA API without VIRTIO_F_ACCESS_PLATFORM) are that we detect that:
>>
>> - secure guest mode is enabled - so we know that since we don't share
>> most memory regular virtio code won't
>> work, even though the buggy hypervisor didn't set VIRTIO_F_ACCESS_PLATFORM
>
> force_dma_unencrypted(&vdev->dev) is IMHO exactly about this.
>
>> - DMA API is giving us addresses that are actually also physical
>> addresses
>
> In case of s390 this is given. I talked with the power people before
> posting this, and they ensured me they can are willing to deal with
> this. I was hoping to talk abut this with the AMD SEV people here (hence
> the cc).
Yes, physical addresses are fine for SEV - the key is that the DMA API is
used so that an address for unencrypted, or shared, memory is returned.
E.g. for a dma_alloc_coherent() call this is an allocation that has had
set_memory_decrypted() called or for a dma_map_page() call this is an
address from SWIOTLB, which was mapped shared during boot, where the data
will be bounce-buffered.
We don't currently support an emulated IOMMU in our SEV guest because that
would require a lot of support in the driver to make IOMMU data available
to the hypervisor (I/O page tables, etc.). We would need hardware support
to really make this work easily in the guest.
Thanks,
Tom
>
>> - Hypervisor is buggy and didn't enable VIRTIO_F_ACCESS_PLATFORM
>>
>
> I don't get this point. The argument where the hypervisor is buggy is a
> bit hard to follow for me. If hypervisor is buggy we have already lost
> anyway most of the time, or?
>
>> I don't see how this patch does this.
>
> I do get your point. I don't know of a good way to check that DMA API
> is giving us addresses that are actually physical addresses, and the
> situation you describe definitely has some risk to it.
>
> Let me comment on other ideas that came up. I would be very happy to go
> with the best one. Thank you very much.
>
> Regards,
> Halil
>
>>
>>
>>> ---
>>> drivers/virtio/virtio_ring.c | 3 +++
>>> 1 file changed, 3 insertions(+)
>>>
>>> diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
>>> index 867c7ebd3f10..fafc8f924955 100644
>>> --- a/drivers/virtio/virtio_ring.c
>>> +++ b/drivers/virtio/virtio_ring.c
>>> @@ -243,6 +243,9 @@ static bool vring_use_dma_api(struct virtio_device *vdev)
>>> if (!virtio_has_iommu_quirk(vdev))
>>> return true;
>>>
>>> + if (force_dma_unencrypted(&vdev->dev))
>>> + return true;
>>> +
>>> /* Otherwise, we are left to guess. */
>>> /*
>>> * In theory, it's possible to have a buggy QEMU-supposed
>>> --
>>> 2.17.1
>>
>
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2020-02-21 16:12 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-20 16:06 [PATCH 0/2] virtio: decouple protected guest RAM form VIRTIO_F_IOMMU_PLATFORM Halil Pasic
2020-02-20 16:06 ` [PATCH 1/2] mm: move force_dma_unencrypted() to mem_encrypt.h Halil Pasic
2020-02-20 16:11 ` Christoph Hellwig
2020-02-20 16:23 ` Christian Borntraeger
2020-02-20 16:31 ` Christoph Hellwig
2020-02-20 17:00 ` Christian Borntraeger
2020-02-21 3:27 ` David Gibson
2020-02-21 13:06 ` Halil Pasic
2020-02-21 15:48 ` Michael S. Tsirkin
2020-02-21 18:07 ` Halil Pasic
2020-02-24 3:33 ` David Gibson
2020-02-24 18:49 ` Halil Pasic
2020-02-25 18:08 ` Cornelia Huck
2020-02-28 0:23 ` David Gibson
2020-02-20 16:06 ` [PATCH 2/2] virtio: let virtio use DMA API when guest RAM is protected Halil Pasic
2020-02-20 16:13 ` Christoph Hellwig
2020-02-21 2:59 ` David Gibson
2020-02-21 3:41 ` Jason Wang
2020-02-21 13:31 ` Halil Pasic
2020-02-21 13:27 ` Halil Pasic
2020-02-21 16:36 ` Christoph Hellwig
2020-02-24 6:50 ` David Gibson
2020-02-24 18:59 ` Halil Pasic
2020-02-21 14:33 ` Halil Pasic
2020-02-21 16:39 ` Christoph Hellwig
2020-02-21 18:16 ` Halil Pasic
2020-02-22 19:07 ` Michael S. Tsirkin
2020-02-24 17:16 ` Christoph Hellwig
2020-10-28 14:24 ` Alexander Graf via iommu
2020-10-28 18:01 ` Michael S. Tsirkin
2020-02-20 20:55 ` Michael S. Tsirkin
2020-02-21 1:17 ` Ram Pai
2020-02-21 3:29 ` David Gibson
2020-02-21 13:12 ` Halil Pasic
2020-02-21 15:39 ` Tom Lendacky [this message]
2020-02-24 6:40 ` David Gibson
2020-02-21 15:56 ` Michael S. Tsirkin
2020-02-21 16:35 ` Christoph Hellwig
2020-02-21 18:03 ` Halil Pasic
2020-02-20 20:48 ` [PATCH 0/2] virtio: decouple protected guest RAM form VIRTIO_F_IOMMU_PLATFORM Michael S. Tsirkin
2020-02-20 21:29 ` Michael S. Tsirkin
2020-02-21 13:37 ` Halil Pasic
2020-02-20 21:33 ` Michael S. Tsirkin
2020-02-21 13:49 ` Halil Pasic
2020-02-21 16:41 ` Christoph Hellwig
2020-02-24 5:44 ` David Gibson
2020-02-21 6:22 ` Jason Wang
2020-02-21 14:56 ` Halil Pasic
2020-02-24 3:38 ` David Gibson
2020-02-24 4:01 ` Jason Wang
2020-02-24 6:06 ` Michael S. Tsirkin
2020-02-24 6:45 ` Jason Wang
2020-02-24 7:48 ` Michael S. Tsirkin
2020-02-24 9:26 ` Jason Wang
2020-02-24 13:40 ` Michael S. Tsirkin
2020-02-25 3:38 ` Jason Wang
2020-02-24 13:56 ` Halil Pasic
2020-02-25 3:30 ` Jason Wang
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=bf0c87a2-adc6-a05a-835f-f98e7cd0cd3b@amd.com \
--to=thomas.lendacky@amd.com \
--cc=borntraeger@de.ibm.com \
--cc=brijesh.singh@amd.com \
--cc=cohuck@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=frankja@linux.ibm.com \
--cc=hch@lst.de \
--cc=iommu@lists.linux-foundation.org \
--cc=jasowang@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=linuxram@us.ibm.com \
--cc=mihajlov@linux.ibm.com \
--cc=mimu@linux.ibm.com \
--cc=mst@redhat.com \
--cc=pasic@linux.ibm.com \
--cc=robin.murphy@arm.com \
--cc=virtualization@lists.linux-foundation.org \
--cc=wei.huang2@amd.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).