[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [vfio-users] Small patch for those that have nVidia cards and want to easily use LibVirt.

I'm on Gentoo so it's easy to add patches to an ebuild -- just drop then in a certain folder and the ebuild adds them so it's just reinstalling. As Alex said and what I said in my email, the custom command line args section is not good to do and his way with the script or my patch avoids that. To me the patch is easier and I have the HyperV stuff in the XML file.


On 17 Mar 2016 17:48, "Alex Williamson" <alex l williamson gmail com> wrote:
On Thu, Mar 17, 2016 at 11:42 AM, Corey Larson <corey eatrunco de> wrote:
On Thu, Mar 17, 2016, at 06:56 AM, Alex Williamson wrote:
There are definitely easier ways to do this than recompiling QEMU.  I favor a wrapper script such that you simply replace the <emulator> in the domain XML with the wrapper and it works automatically, something like:
$ cat /usr/libexec/qemu-kvm-hv-vendor 
exec /usr/bin/qemu-kvm \
`echo "\$@" | sed 's|hv_time|hv_time,hv_vendor_id=KeenlyKVM|g'`
vfio-users mailing list
I was going to be trying this approach soon: https://libvirt.org/drvqemu.html#qemucommand. Is there any reason to prefer this scripting approach over the XML approach?

<qemu:arg> is inherently evil.  You end up specifying two -cpu args, the one libvirt generates and the one you specific in the qemucommand (which is opaque to libvirt) and you hope that qemu selects the one you want.  Using virt-manager to fiddle with cpu options might become meaningless, we don't know whether qemu merges all cpu options together or only takes the last (we could look, but it's probably unexpected and therefore undefined and could change at any release). 

vfio-users mailing list
vfio-users redhat com

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]