Re: "xl pci-list-assignable" command not implemented (Was: Re: [xen-unstable test] 12889: regressions - FAIL)

[Ian Jackson <Ian.Jackson@eu.citrix.com>]

Ian Campbell writes ("Re: [Xen-devel] "xl pci-list-assignable" command not implemented (Was: Re: [xen-unstable test] 12889: regressions - FAIL)"):
> On Wed, 2012-05-16 at 11:24 +0100, George Dunlap wrote:
> > On Wed, May 16, 2012 at 9:34 AM, Ian Campbell <Ian.Campbell@citrix.com> wrote:
> > > I guess we could make the test harness detect which to use, or perhaps
> > > we should just put in a hidden alias command?
> > 
> > Why do we need to detect which to use?
> 
> For the benefit of the bisector when crossing this changeset.
> 
> >   I pci-list-assignable was
> > introduced in the 4.2 dev cycle, I believe (which is why I wasn't
> > worried about changing the name); was it backported to 4.1?

In the interests of expedience, I have simply changed the tester to
use the new name (well, pushed a change to the tester's staging tree
to do so).  Bisection will not work if the tester tries to bisect the
pcipt job but I think it's unlikely we'll have another nonobvious
bisectable regression given that the /actual/ test never passes.

Ian.