From: Alex Williamson <alex.williamson@redhat.com>
To: Nan Xiao <xiaonan830818@gmail.com>
Cc: kvm@vger.kernel.org
Subject: Re: A question about "CONFIG_KVM_DEVICE_ASSIGNMENT" configuration
Date: Thu, 10 Sep 2015 18:18:07 -0600 [thread overview]
Message-ID: <1441930687.20355.658.camel@redhat.com> (raw)
In-Reply-To: <CA+MhoaNfBO=y-5VMz7gNsZc0RUp1YSuEYHOjGwP=DXj0wbYkOw@mail.gmail.com>
On Fri, 2015-09-11 at 07:58 +0800, Nan Xiao wrote:
> Hi Alex,
>
> Thanks very much for your response! so for assigning PCI device in QEMU.
> The following method is deprecated:
>
> qemu-system-x86_64 ... -device pci-assign,host=0000:21:00.1
>
> I should use this one instead:
>
> qemu-system-x86_64 ... -device vfio-pci,host=0000:21:00.1
>
> right?
Yes, and the device needs to be bound to the vfio_pci driver in the host
rather than pci-stub (and you may or may not need to care about IOMMU
groups[1]). Thanks,
Alex
[1] http://vfio.blogspot.com/2014/08/iommu-groups-inside-and-out.html
> On Thu, Sep 10, 2015 at 11:08 PM, Alex Williamson
> <alex.williamson@redhat.com> wrote:
> > On Thu, 2015-09-10 at 18:22 +0800, Nan Xiao wrote:
> >> Hi all,
> >>
> >> When building kernel, it prompts "CONFIG_KVM_DEVICE_ASSIGNMENT" is "deprecated".
> >> But it is still used in kernel code. E.g.:
> >> "kvm_vm_ioctl_check_extension" function:
> >> {
> >> ...
> >> #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
> >> case KVM_CAP_ASSIGN_DEV_IRQ:
> >> case KVM_CAP_PCI_2_3:
> >> #endif
> >> r = 1;
> >> break;
> >> ...
> >> #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
> >> case KVM_CAP_IOMMU:
> >> r = iommu_present(&pci_bus_type);
> >> break;
> >> #endif
> >> ...
> >> }
> >>
> >> If not configure this option, the following code will execute failed:
> >>
> >> ret = ioctl(dev, KVM_CHECK_EXTENSION, KVM_CAP_IOMMU);
> >>
> >> So does it mean to use KVM assigned device feature, the
> >> "CONFIG_KVM_DEVICE_ASSIGNMENT"
> >> is not "deprecated"?
> >
> > Legacy KVM device assignment is deprecated, it is fully replaced by
> > VFIO-based device assignment. The intention is to deprecate legacy KVM
> > device assignment now, so all users can transition away from it and at
> > some point remove it from the kernel. When using QEMU and libvirt, the
> > default is already to use VFIO instead of legacy KVM device assignment.
> > Thanks,
> >
> > Alex
> >
> >
> >
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2015-09-11 0:18 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-10 10:22 A question about "CONFIG_KVM_DEVICE_ASSIGNMENT" configuration Nan Xiao
2015-09-10 15:08 ` Alex Williamson
2015-09-10 23:58 ` Nan Xiao
2015-09-11 0:18 ` Alex Williamson [this message]
2015-09-11 0:42 ` Nan Xiao
2015-09-11 0:55 ` Chen, Tiejun
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=1441930687.20355.658.camel@redhat.com \
--to=alex.williamson@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=xiaonan830818@gmail.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 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.