On Mon, 21 Aug 2017 14:00:15 +0200 Halil Pasic <pa...@linux.vnet.ibm.com> wrote:
> On 08/21/2017 11:16 AM, Cornelia Huck wrote: > > If we don't provide pci, we cannot have a pci device for which we > > have to translate to adapter routes: just return -ENODEV. > > > > Signed-off-by: Cornelia Huck <coh...@redhat.com> > > --- > > target/s390x/kvm.c | 6 ++++++ > > 1 file changed, 6 insertions(+) > > > > diff --git a/target/s390x/kvm.c b/target/s390x/kvm.c > > index 9de165d8b1..d8db1cbf6e 100644 > > --- a/target/s390x/kvm.c > > +++ b/target/s390x/kvm.c > > @@ -2424,6 +2424,12 @@ int kvm_arch_fixup_msi_route(struct > > kvm_irq_routing_entry *route, > > uint32_t idx = data >> ZPCI_MSI_VEC_BITS; > > uint32_t vec = data & ZPCI_MSI_VEC_MASK; > > > > + if (!s390_has_feat(S390_FEAT_ZPCI)) { > > + /* How can we get here without pci enabled? */ > > + g_assert(false); > > You don't tell us about the g_assert in the commit message. > Do you expect G_DISABLE_ASSERT being defined for production > builds. I've grepped for G_DISABLE_ASSERT and found nothing. AFAIK this is set by distribution builds. I've also noticed that mingw builds treat (g_)assert() as if code flow continues, but I don't know whether asserts do anything there at all. > > And why g_assert over assert (again no guidance in HACKING > mostly asking for my own learning)? I do recall a recent(ish) discussion, but not the details. Anyway, using glib interfaces seems more consistent. > > Other that that LGTM. Thanks! > > > > + return -ENODEV; > > + } > > + > > pbdev = s390_pci_find_dev_by_idx(s390_get_phb(), idx); > > if (!pbdev) { > > DPRINTF("add_msi_route no dev\n"); > > >