From: Haggai Eran <haggaie@mellanox.com>
To: Alex Williamson <alex.williamson@redhat.com>,
Ilya Lesokhin <ilyal@mellanox.com>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"linux-pci@vger.kernel.org" <linux-pci@vger.kernel.org>
Cc: "bhelgaas@google.com" <bhelgaas@google.com>,
Noa Osherovich <noaos@mellanox.com>,
Or Gerlitz <ogerlitz@mellanox.com>,
Liran Liss <liranl@mellanox.com>
Subject: Re: [RFC 0/2] VFIO SRIOV support
Date: Tue, 5 Jan 2016 13:22:31 +0200 [thread overview]
Message-ID: <568BA777.6090707@mellanox.com> (raw)
In-Reply-To: <1450965103.2950.92.camel@redhat.com>
On 24/12/2015 15:51, Alex Williamson wrote:
> No. A privileged entity needs to grant a user ownership of a group and
> sufficient locked memory limits to make it useful, but then use of the
> group does not require root permission.
So we're thinking how we can force the VFs in these cases to be in the same
IOMMU group with the PF, and make sure it is vfio-pci that probes them. We
thought about the following:
We could add a flag to pci_dev->dev_flags on the PF, that says that the PF's
VFs must be in the same IOMMU group with it. Modify
iommu_group_get_for_pci_dev() so that it will return the PFs group for VFs
whose PF has that flag set.
In the vfio_group_nb_add_dev() function set driver_override to "vfio-pci" for
PCI devices that are added to a live group. That would prevent the host from
probing these devices with the default driver.
What do you think?
Regards,
Haggai and Ilya
prev parent reply other threads:[~2016-01-05 11:22 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-22 13:42 [RFC 0/2] VFIO SRIOV support Ilya Lesokhin
2015-12-22 13:42 ` [RFC 1/2] PCI: Expose iov_set_numvfs and iov_resource_size for modules Ilya Lesokhin
2015-12-22 13:42 ` [RFC 2/2] VFIO: Add support for SRIOV extended capablity Ilya Lesokhin
2015-12-22 15:35 ` [RFC 0/2] VFIO SRIOV support Alex Williamson
2015-12-23 7:43 ` Ilya Lesokhin
2015-12-23 16:28 ` Alex Williamson
2015-12-24 7:22 ` Ilya Lesokhin
2015-12-24 13:51 ` Alex Williamson
2016-01-05 11:22 ` Haggai Eran [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=568BA777.6090707@mellanox.com \
--to=haggaie@mellanox.com \
--cc=alex.williamson@redhat.com \
--cc=bhelgaas@google.com \
--cc=ilyal@mellanox.com \
--cc=kvm@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=liranl@mellanox.com \
--cc=noaos@mellanox.com \
--cc=ogerlitz@mellanox.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).