From: Christoph Hellwig <hch@lst.de>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>, Christoph Hellwig <hch@lst.de>,
Joerg Roedel <joro@8bytes.org>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
Jens Axboe <axboe@kernel.dk>,
virtualization@lists.linux-foundation.org,
linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
iommu@lists.linux-foundation.org, jfehlig@suse.com,
jon.grimm@amd.com, brijesh.singh@amd.com
Subject: Re: [PATCH 0/3] Fix virtio-blk issue with SWIOTLB
Date: Mon, 14 Jan 2019 21:19:35 +0100 [thread overview]
Message-ID: <20190114201935.GA10781@lst.de> (raw)
In-Reply-To: <20190114131114-mutt-send-email-mst@kernel.org>
On Mon, Jan 14, 2019 at 01:20:45PM -0500, Michael S. Tsirkin wrote:
> I don't think so - the issue is really that DMA API does not yet handle
> the SEV case 100% correctly. I suspect passthrough devices would have
> the same issue.
The DMA API handles the SEV case perfectly. Its just that virtio_blk
supports huge segments that virtio does not generally support, but that
is not related to SEV in any way.
> In fact whoever sets IOMMU_PLATFORM is completely unaffected by
> Christoph's pet peeve.
No, the above happens only when we set IOMMU_PLATFORM.
> Christoph is saying that !IOMMU_PLATFORM devices should hide the
> compatibility code in a special per-device DMA API implementation.
> Which would be fine especially if we can manage not to introduce a bunch
> of indirect calls all over the place and hurt performance. It's just
> that the benefit is unlikely to be big (e.g. we can't also get rid of
> the virtio specific memory barriers) so no one was motivated enough to
> work on it.
No. The problem is that we still haven't fully specified what
IOMMU_PLATFORM and !IOMMU_PLATFORM actually mean. Your
"ACCESS_PLATFORM/ORDER_PLATFORM" commit in the virtio-spec repo
improves it a little bit, but it is still far from enough.
As a start VIRTIO_F_ORDER_PLATFORM and VIRTIO_F_ACCESS_PLATFORM
absolutely MUST be set for hardware implementations. Otherwise said
hardware has no chance of working on anything but the most x86-like
systems.
Second software implementations SHOULD set VIRTIO_F_ACCESS_PLATFORM,
because otherwise we can't add proper handling for things like SEV or
the IBM "secure hypervisor" thing.
Last but not least a lot of wording outside the area describing these
flags really needs some major updates in terms of DMA access.
next prev parent reply other threads:[~2019-01-14 20:19 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-10 13:44 [PATCH 0/3] Fix virtio-blk issue with SWIOTLB Joerg Roedel
2019-01-10 13:44 ` [PATCH 1/3] swiotlb: Export maximum allocation size Joerg Roedel
2019-01-10 17:02 ` Konrad Rzeszutek Wilk
2019-01-11 9:12 ` Joerg Roedel
2019-01-14 20:49 ` Konrad Rzeszutek Wilk
2019-01-14 21:59 ` Michael S. Tsirkin
2019-01-15 13:05 ` Christoph Hellwig
2019-01-10 13:44 ` [PATCH 2/3] virtio: Introduce virtio_max_dma_size() Joerg Roedel
2019-01-10 13:44 ` [PATCH 3/3] virtio-blk: Consider virtio_max_dma_size() for maximum segment size Joerg Roedel
2019-01-10 13:59 ` [PATCH 0/3] Fix virtio-blk issue with SWIOTLB Christoph Hellwig
2019-01-10 14:26 ` Joerg Roedel
2019-01-11 3:29 ` Jason Wang
2019-01-11 9:15 ` Joerg Roedel
2019-01-14 9:41 ` Jason Wang
2019-01-14 9:50 ` Christoph Hellwig
2019-01-14 12:41 ` Jason Wang
2019-01-14 18:20 ` Michael S. Tsirkin
2019-01-14 19:09 ` Singh, Brijesh
2019-01-14 19:12 ` Robin Murphy
2019-01-14 20:22 ` Christoph Hellwig
2019-01-14 20:29 ` Michael S. Tsirkin
2019-01-14 20:19 ` Christoph Hellwig [this message]
2019-01-14 20:48 ` Michael S. Tsirkin
2019-01-15 13:09 ` Christoph Hellwig
2019-01-15 8:37 ` Joerg Roedel
2019-01-15 13:20 ` Christoph Hellwig
2019-01-16 14:16 ` Michael S. Tsirkin
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=20190114201935.GA10781@lst.de \
--to=hch@lst.de \
--cc=axboe@kernel.dk \
--cc=brijesh.singh@amd.com \
--cc=iommu@lists.linux-foundation.org \
--cc=jasowang@redhat.com \
--cc=jfehlig@suse.com \
--cc=jon.grimm@amd.com \
--cc=joro@8bytes.org \
--cc=konrad.wilk@oracle.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@redhat.com \
--cc=virtualization@lists.linux-foundation.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).