[vfio-users] CPU pinning with cgroups

Tue Sep 8 15:44:30 UTC 2015

You can redirect qemu monitor by qemu commandline options in libvirt. But
wasn't there a built-in function to do that?
Also, try using virsh, i'm sure you can reboot VMs with it.
On Sep 8, 2015 6:40 PM, "Quentin Deldycke" <quentindeldycke at gmail.com>
wrote:

> While using directly the monitor you can plug / unplug usb from the vm
> using a simple shell command...
> Also, it is interesting in case your vm crash to be able to take back
> control to host (using ssh + shell command from phone)
>
> --
> Deldycke Quentin
>
>
> On 8 September 2015 at 17:01, Will Marler <will at wmarler.com> wrote:
>
>> Ha! Interesting. Haven't tried that.
>>
>> On Tue, Sep 8, 2015 at 8:56 AM, Blank Field <ihatethisfield at gmail.com>
>> wrote:
>>
>>> ...but wait, there's more!
>>> If you unplug the p-t'd device from the host while VM is offline, the VM
>>> will fail to boot.
>>> There's still a lot of work to be done with USB.
>>> On Sep 8, 2015 5:51 PM, "Will Marler" <will at wmarler.com> wrote:
>>>
>>>> USB pass-through works quite well with libvirt/virt-manager now; I've
>>>> used it for mouse, keyboard, two different audio devices, while the guest
>>>> was running. There's one feature request that I haven't gotten around to
>>>> submitting; have an "enable/disable" checkbox in the virt-manager gui (I
>>>> will not always want to p-t a USB device, but I will want to frequently
>>>> enough that deleting it & re-adding it each time gets on my nerves).
>>>>
>>>> On Tue, Sep 8, 2015 at 12:15 AM, Quentin Deldycke <
>>>> quentindeldycke at gmail.com> wrote:
>>>>
>>>>> My main problem with libvirt was:
>>>>>
>>>>>
>>>>>    - no ovmf support at the time i created my vm on debian
>>>>>    - No correct usb passthrough
>>>>>
>>>>>
>>>>> The qemu monitor support the device_add command. Which let me bind /
>>>>> unbind usb devices at runtime, without limit of number.
>>>>>
>>>>> I really love libvirt at my job, but for this particular case, i found
>>>>> it less interesting..
>>>>>
>>>>>
>>>>> --
>>>>> Deldycke Quentin
>>>>>
>>>>>
>>>>> On 8 September 2015 at 06:23, Will Marler <will at wmarler.com> wrote:
>>>>>
>>>>>> Not to mention, you can still see the actual qemu invocation script
>>>>>> in the logs when you use libvirt, in case you *really* want to (though a
>>>>>> pastebin share of your libvirt xml config is better, imo ...)
>>>>>>
>>>>>> On Mon, Sep 7, 2015 at 10:15 PM, Jon Panozzo <
>>>>>> jonp at lime-technology.com> wrote:
>>>>>>
>>>>>>> Just to second Alex's point here, I see a lot of folks doing QEMU
>>>>>>> the old fashioned way, which is to say, manual invocation.  Libvirt really
>>>>>>> is an amazing management solution and makes things so much easier.  Lots of
>>>>>>> things can go awry if you're invoking QEMU manually, and libvirt can
>>>>>>> prevent that, as well as automate many other aspects of management that
>>>>>>> manual QEMU invocation cannot.
>>>>>>> On Sep 7, 2015 3:20 PM, "Alex Williamson" <
>>>>>>> alex.l.williamson at gmail.com> wrote:
>>>>>>>
>>>>>>>> On Sun, Sep 6, 2015 at 7:19 PM, Garrett Powell <
>>>>>>>> garretttracypowell at gmail.com> wrote:
>>>>>>>>
>>>>>>>>> I've been having trouble with my guest freezing up, which I'm told
>>>>>>>>> is due to a lack of CPU pinning. I'm currently trying to set up CPU pinning
>>>>>>>>> with cgroups (specifically cpuset), but the freezing isn't going away.
>>>>>>>>> Could anyone tell me what I'm doing wrong? I'm specifically referring to
>>>>>>>>> the *cgcreate*, *cgset *and* cgexec* lines. Here's my launch
>>>>>>>>> script:
>>>>>>>>>
>>>>>>>>> #!/bin/bash
>>>>>>>>>
>>>>>>>>> export QEMU_AUDIO_DRV=pa
>>>>>>>>> export QEMU_PA_SERVER=localhost
>>>>>>>>> export PULSE_SERVER=localhost
>>>>>>>>>
>>>>>>>>> vfio-bind 0000:06:00.0
>>>>>>>>>
>>>>>>>>> synergys --daemon --config /etc/synergy.conf
>>>>>>>>>
>>>>>>>>> for i in {0..3}; do
>>>>>>>>>     echo performance >
>>>>>>>>> /sys/devices/system/cpu/cpu${i}/cpufreq/scaling_governor
>>>>>>>>> done
>>>>>>>>>
>>>>>>>>> echo 5120 > /proc/sys/vm/nr_hugepages
>>>>>>>>>
>>>>>>>>> cgcreate -g cpuset:kvm
>>>>>>>>> cgset -r cpuset.cpus=0-3 -r cpuset.mems=0 -r
>>>>>>>>> cpuset.cpu_exclusive=1 kvm
>>>>>>>>> cgexec -g cpuset:kvm qemu-system-x86_64 \
>>>>>>>>>
>>>>>>>>
>>>>>>>> Ok, so you've change to pCPUs 0-3 instead of 4-7, but doesn't this
>>>>>>>> have the same problem as your previous attempt with taskset?  You're using
>>>>>>>> a different tool, but afaict you're still not doing 1:1 pinning of vCPUs,
>>>>>>>> you're pining them to a set of pCPUs.  Taskset can work, but you need to
>>>>>>>> get the thread ID for each vCPU thread and pin it.  The problem is not the
>>>>>>>> tool you're using but how you're using it, to set the entire process and
>>>>>>>> all the threads it will launch to a range of CPUs rather than pulling out
>>>>>>>> the vCPU threads and individually pinning them.  There's this really great
>>>>>>>> tool that already does this, it's called libvirt.  Thanks,
>>>>>>>>
>>>>>>>> Alex
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> vfio-users mailing list
>>>>>>>> vfio-users at redhat.com
>>>>>>>> https://www.redhat.com/mailman/listinfo/vfio-users
>>>>>>>>
>>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> vfio-users mailing list
>>>>>>> vfio-users at redhat.com
>>>>>>> https://www.redhat.com/mailman/listinfo/vfio-users
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> vfio-users mailing list
>>>>>> vfio-users at redhat.com
>>>>>> https://www.redhat.com/mailman/listinfo/vfio-users
>>>>>>
>>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> vfio-users mailing list
>>>> vfio-users at redhat.com
>>>> https://www.redhat.com/mailman/listinfo/vfio-users
>>>>
>>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/vfio-users/attachments/20150908/7d0ce442/attachment.htm>