On Mon, 2018-03-05 at 16:58 +0100, Andrea Bolognani wrote:
I'm actually fairly confident we could drop that hunk with no ill effect, because the spapr-pci-host-bridge must have existed even before we started checking for it, even if it might possibly have been, at some point, not user-instantiable; however, the capabilities data for ppc64 in our test suite doesn't go far back enough to prove this is the case, so special-casing it seems like the safest option.
I'll try to see if I can build QEMU 1.2.0 (that's the earliest we claim support for, right?) on ppc64 with a reasonable amount of work, and if my hunch is correct I'll post a follow-up cleanup patch that gets rid of it.
It's actually 0.12, not 1.2.0. Whoops. According to my research, the pSeries machine type was introduced in 0.15 but didn't have PCI support until 1.0, by which point the spapr-pci-host-bridge device had been introduced too. tl;dr the hunk in question can safely be dropped. I'll respin. -- Andrea Bolognani / Red Hat / Virtualization