Re: [Qemu-devel] [PATCH v4 03/10] tests: Add vm test lib

[Fam Zheng <famz@redhat.com>]

On Fri, 09/01 16:29, Philippe Mathieu-Daudé wrote:
> On 08/30/2017 12:34 AM, Fam Zheng wrote:
> > On Tue, 08/29 14:34, Philippe Mathieu-Daudé wrote:
> > > > +        self._args = [ \
> > > > +            "-nodefaults", "-m", "2G",
> > > > +            "-cpu", "host",
> > > > +            "-netdev", "user,id=vnet,hostfwd=:0.0.0.0:0-:22",
> > > > +            "-device", "virtio-net-pci,netdev=vnet",
> > > > +            "-vnc", ":0,to=20",
> > > > +            "-serial", "file:%s" % os.path.join(self._tmpdir, "serial.out")]
> > > > +        if vcpus:
> > > > +            self._args += ["-smp", str(vcpus)]
> > > 
> > > What about enabling mttcg which isn't default?
> > > 
> > >              self._args += ["--accel", "tcg,thread=multi"]
> > 
> > Any specific reason to enable it? I think it is not available on older QEMU.
> 
> Neither is dynamic portfwd :)

I figured, but portfwd is strongly justified, whereas ...

> 
> I see 2 reasons:
> - faster test

Any data? And if it is noticably faster, I doubt anyone is going to actually use
it, because it probably take a whole day to run one build.

> - cover mttcg

Testing mttcg is good, but we don't want to test mttcg and building at the same
time. Again, it can take a whole day.

But anyway --accel and any other options should be possible to get passed as
command line options.

Fam

> 
> > 
> > > 
> > > > +        if os.access("/dev/kvm", os.R_OK | os.W_OK):
> > > > +            self._args += ["-enable-kvm"]
> > > > +        else:
> > > > +            logging.info("KVM not available, not using -enable-kvm")
> > > > +        self._data_args = []
> > > [...]
> > 
> > Fam
> > 
>