On 03/28/2018 10:06 AM, Andrea Bolognani wrote:
I haven't been able to come up with a single scenario in which the code in question would be executed; even if there was one, it would be due to the user specifying a *partial* PCI topology in the guest XML, which is of course entirely unsupportable and thus providing even the slightest hint that doing so is in any way a good idea is actively harmful.
Signed-off-by: Andrea Bolognani <abologna@redhat.com> --- src/conf/domain_addr.c | 9 --------- 1 file changed, 9 deletions(-)
Since Laine added this code - perhaps calling his name out on the CC list will allow him to appear and answer the question. Although it could take 3 such utterances (e.g. beetlejuice)... Personally, it seems reasonable and you didn't have to change any test output, but PCI connection requirements are black boxes to me. I know I cannot plug my US plug into the CZ outlets, but when it comes to PCI connection rules - I'm glad someone else knows them! A soft and mostly unqualified, Reviewed-by: John Ferlan <jferlan@redhat.com> John
diff --git a/src/conf/domain_addr.c b/src/conf/domain_addr.c index 0c914fe25c..18b6f8d588 100644 --- a/src/conf/domain_addr.c +++ b/src/conf/domain_addr.c @@ -447,15 +447,6 @@ virDomainPCIAddressSetGrow(virDomainPCIAddressSetPtr addrs, addr->bus++; } } - } else if (flags & VIR_PCI_CONNECT_TYPE_PCI_BRIDGE && - addrs->buses[0].model == VIR_DOMAIN_CONTROLLER_MODEL_PCIE_ROOT) { - /* NB: if the root bus is pci-root, and we couldn't find an - * open place to connect a pci-bridge, then there is nothing - * we can do (since the only way to gain a new slot that - * accepts a pci-bridge is to add *a pci-bridge* (which is the - * reason we're here in the first place!) - */ - model = VIR_DOMAIN_CONTROLLER_MODEL_DMI_TO_PCI_BRIDGE; } else if (flags & (VIR_PCI_CONNECT_TYPE_PCIE_DEVICE | VIR_PCI_CONNECT_TYPE_PCIE_SWITCH_UPSTREAM_PORT)) { model = VIR_DOMAIN_CONTROLLER_MODEL_PCIE_ROOT_PORT;