Message ID | 871ti57y6f.fsf@abhimanyu.in.ibm.com |
---|---|
State | New |
Headers | show |
On Mon, May 25, 2015 at 03:53:52PM +0530, Nikunj A Dadhania wrote: > Alexey Kardashevskiy <aik@ozlabs.ru> writes: > > On 05/25/2015 02:45 PM, Nikunj A Dadhania wrote: > >> Alexey Kardashevskiy <aik@ozlabs.ru> writes: [snip] > >>> Can we also add a hack here to scan for the "qemu,phb-enumerated" string in > >>> the SLOF bin image? > >> > >> Really ? That would be ugly. > > > > > > Well, chances that the binary image will have this line by accident are zero. > > > > And I spent quite some time debugging SRIOV + VFIO when I realized that > > SLOF is old on the test machine where others used to debug too. It would be > > really nice to have a warning that something is wrong. May be extend > > "client-architecture-support" somehow or have some release/date signature > > in known place in SLOF... Thomas (?) also asked for this :) > > Sure, I can work on this. I would not recommend grepping though. Yeah, don't do this. The ugliness is not worth it.
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c index d11e2ab..5737839 100644 --- a/hw/ppc/spapr_pci.c +++ b/hw/ppc/spapr_pci.c @@ -1103,6 +1103,10 @@ static void spapr_phb_add_pci_device(sPAPRDRConnector *drc, error_setg(errp, "Failed to create pci child device tree node"); goto out; } + } else { + int slot = PCI_SLOT(pdev->devfn); + int func = PCI_FUNC(pdev->devfn); + fprintf(stderr, "non-hotplug - pci@%d,%d\n", slot, func); } $ ./ppc64-softmmu/qemu-system-ppc64 -machine pseries -m 2G -serial stdio