On Fri, Sep 28, 2012 at 09:20:01AM -0600, Eric Blake wrote:
On 09/28/2012 08:58 AM, Daniel P. Berrange wrote:
> From: "Daniel P. Berrange" <berrange(a)redhat.com>
>
> Start a QEMU process using
>
> $QEMU -S -no-user-config -nodefaults \
> -nographic -M none -qmp unix:/some/path,server,nowait
>
> and talk QMP over stdio to discover what capabilities the
> binary supports. This works for QEMU 1.2.0 or later and
> for older QEMU automatically fallback to the old approach
> of parsing -help and related command line args.
>
> Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com>
> ---
> src/qemu/qemu_capabilities.c | 394 +++++++++++++++++++++++++++++++++++++++----
> 1 file changed, 358 insertions(+), 36 deletions(-)
>
> +
> +/* Capabilities that we assume are always enabled
> + * for QEMU >= 1.2.0
> + */
> +static void
> +qemuCapsInitQMPBasic(qemuCapsPtr caps)
> +{
> + qemuCapsSet(caps, QEMU_CAPS_VNC_COLON);
> + qemuCapsSet(caps, QEMU_CAPS_NO_REBOOT);
I think qemuCapsSetList() would be nicer than setting one bit at a time,
but that doesn't change semantics.
> +
> +static int
> +qemuCapsInitQMP(qemuCapsPtr caps)
> +{
> + int ret = -1;
> + virCommandPtr cmd = NULL;
> + qemuMonitorPtr mon = NULL;
> + int major, minor, micro;
> + char *package;
> + int status = 0;
> + virDomainChrSourceDef config;
> +
> + memset(&config, 0, sizeof(config));
> + config.type = VIR_DOMAIN_CHR_TYPE_UNIX;
> + config.data.nix.path = (char*)"/tmp/qemu.sock";
This does not seem like a safe file name (it is easily guessable).
Should we instead be generating something under /var/run?
Of course it isn't :-( I blame the end of week exhaustion. Locally
I have an addition to make it use driver->libDir, as we do for normal
monitor sockets for actual VMs.
Daniel
--
|:
http://berrange.com -o-
http://www.flickr.com/photos/dberrange/ :|
|:
http://libvirt.org -o-
http://virt-manager.org :|
|:
http://autobuild.org -o-
http://search.cpan.org/~danberr/ :|
|:
http://entangle-photo.org -o-
http://live.gnome.org/gtk-vnc :|