Re: [Qemu-devel] [PATCH RFC] accel: default to an actually available accelerator

[Kevin Wolf]

Am 07.09.2017 um 10:25 hat Cornelia Huck geschrieben:
> On Thu, 7 Sep 2017 10:14:27 +0200
> Thomas Huth <address@hidden> wrote:
> 
> > On 07.09.2017 10:11, Kevin Wolf wrote:
> > [...]
> > > But the real reason why I'm replying: Should we add changing the default
> > > to "kvm:tcg" to the list of planned 3.0 changes? I am part of the group
> > > that intentionally uses TCG occasionally, but I think the majority of
> > > users wants to use KVM (or whatever the fastest option is on their
> > > system) whenever it is available.  
> > 
> > If you consider how often people are getting this wrong (they want to
> > use KVM but end up with TCG in the first try), I think that's a good idea.
> 
> Agreed.
> 
> I'm wondering what that means for our tests, though. Some of them work
> slightly different under tcg or kvm (cf. iotest 186, as referenced in
> the original mail), and sometimes you'll probably explicitly want to
> exercise tcg. That does not speak against the change, but we probably
> need to look at what we want in more detail.

This is a bug in test 186, and probably 172, too. Normally, we use the
options from ./common:

    export QEMU_OPTIONS="-nodefaults -machine accel=qtest"

However, these two test cases overwrite QEMU_OPTIONS and neglect to add
a '-machine accel=qtest' option manually.

Kevin