Re: [Qemu-devel] Advice on some configuration parameters

[Fabio Fantoni]

Il 03/03/2013 11:27, Alon Levy ha scritto:
> > Il 01/03/2013 13:45, Paolo Bonzini ha scritto:
> > > Il 12/02/2013 14:05, Fabio Fantoni ha scritto:
> > > > I'm making patches to enable some qemu upstream features in xen
> > > > that are
> > > > missing in libxl.
> > > > I'm trying to do it just by giving arguments to qemu, and I want
> > > > them to
> > > > be dynamic (e.g. without physical addresses if possible) and
> > > > concise.
> > > They should use physical addresses, otherwise changing the VM
> > > hardware
> > > in trivial ways may cause a waterfall effect and cause Windows to
> > > reactivate.
> > >
> > > > If I try to set up virt-manager so that it doesn't point to a
> > > > file, it
> > > > uses this configuration:
> > > >
> > > > -device ich9-usb-ehci1,id=usb,bus=pci.0,addr=0x5.0x7
> > > > -device
> > > > ich9-usb-uhci1,masterbus=usb.0,firstport=0,bus=pci.0,multifunction=on,addr=0x5
> > > >
> > > > -device
> > > > ich9-usb-uhci2,masterbus=usb.0,firstport=2,bus=pci.0,addr=0x5.0x1
> > > > -device
> > > > ich9-usb-uhci3,masterbus=usb.0,firstport=4,bus=pci.0,addr=0x5.0x2
> > > >
> > > > This seems to differ from the configuration file, e.g. you have
> > > > multifunzion=on only on the first uhci
> > > That doesn't matter.
> > >
> > > > , besides I can't understand how many physical ports are defined.
> > > There are 6 ports, controlled by either the EHCI for USB 2.0
> > > devices
> > > (the EHCI controls 6 ports) or by one of the three UHCI for USB 1.1
> > > ports (one UHCI has 2 ports).
> > >
> > > > I tried setting 9 usbredirection
> > > > channels in virt-manager but that configuration doesn't change and
> > > > no
> > > > errors are reported, but I doubt that physical ports are being
> > > > define
> > > > dynamically.
> > > QEMU creates hubs automatically I think in that case.
> > >
> > > > About spice vdagent at the moment I'm using this configuration
> > > > that
> > > > seems to work:
> > > > -device virtio-serial -chardev spicevmc,id=vdagent,name=vdagent
> > > > -device
> > > > virtserialport,chardev=vdagent,name=com.redhat.spice.0
> > > >
> > > > But I noticed that virt-manager use this slightly different one:
> > > > -chardev pty,id=charserial0 -device
> > > > isa-serial,chardev=charserial0,id=serial0 -chardev
> > > > spicevmc,id=charchannel0,name=vdagent -device
> > > > virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel0,id=channel0,name=com.redhat.spice.0
> > > >
> > > > Are there any problem with the more concise one I'm using?
> > > It is the same.
> > >
> > > Paolo
> > Thanks for reply.
> > I have another questions:
> > What exactly does the -usb parameter that seems to be necessary for
> > usb
> > passthrough?
> It is shorthand for the appropriate "-device" to create the usb host (i.e. uhci 
> etc.)

Thanks for the reply, I tried -usb and it seems to create an usb1 
controller with 2 ports and not an enhanced usb2 with 6 ports.

> > I mean usb devices passed through host (with -usbdevice).
> > Is the number of such passthrough usb devices limiting the number of
> > available channel for usb redirection devices?
> AFAIR yes.
>
> >

smime.p7s
Description: Firma crittografica S/MIME