Message ID | 20210722090009.383575-1-marcel@redhat.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | hw/pcie-root-port: Fix hotplug for PCI devices requiring IO | expand |
On Thu, 22 Jul 2021 12:00:09 +0300 Marcel Apfelbaum <marcel.apfelbaum@gmail.com> wrote: > Q35 has now ACPI hotplug enabled by default for PCI(e) devices. > As opposed to native PCIe hotplug, some guests will not assign could be 'some guests' be more specific, preferably with a way to reproduce issue. > IO range to pcie-root-ports not supporting native hotplug, > resulting into a regression. > > Fix it by setting the "reserve-io" hint capability of the > pcie-root-ports so the firmware will allocate the IO range instead. > > Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> other than commit message nit, looks fine to me Acked-by: Igor Mammedov <imammedo@redhat.com> > --- > hw/pci-bridge/gen_pcie_root_port.c | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/hw/pci-bridge/gen_pcie_root_port.c b/hw/pci-bridge/gen_pcie_root_port.c > index ec9907917e..20099a8ae3 100644 > --- a/hw/pci-bridge/gen_pcie_root_port.c > +++ b/hw/pci-bridge/gen_pcie_root_port.c > @@ -28,6 +28,7 @@ OBJECT_DECLARE_SIMPLE_TYPE(GenPCIERootPort, GEN_PCIE_ROOT_PORT) > (GEN_PCIE_ROOT_PORT_AER_OFFSET + PCI_ERR_SIZEOF) > > #define GEN_PCIE_ROOT_PORT_MSIX_NR_VECTOR 1 > +#define GEN_PCIE_ROOT_DEFAULT_IO_RANGE 4096 > > struct GenPCIERootPort { > /*< private >*/ > @@ -75,6 +76,7 @@ static bool gen_rp_test_migrate_msix(void *opaque, int version_id) > static void gen_rp_realize(DeviceState *dev, Error **errp) > { > PCIDevice *d = PCI_DEVICE(dev); > + PCIESlot *s = PCIE_SLOT(d); > GenPCIERootPort *grp = GEN_PCIE_ROOT_PORT(d); > PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(d); > Error *local_err = NULL; > @@ -85,6 +87,9 @@ static void gen_rp_realize(DeviceState *dev, Error **errp) > return; > } > > + if (grp->res_reserve.io == -1 && s->hotplug && !s->native_hotplug) { > + grp->res_reserve.io = GEN_PCIE_ROOT_DEFAULT_IO_RANGE; > + } > int rc = pci_bridge_qemu_reserve_cap_init(d, 0, > grp->res_reserve, errp); >
diff --git a/hw/pci-bridge/gen_pcie_root_port.c b/hw/pci-bridge/gen_pcie_root_port.c index ec9907917e..20099a8ae3 100644 --- a/hw/pci-bridge/gen_pcie_root_port.c +++ b/hw/pci-bridge/gen_pcie_root_port.c @@ -28,6 +28,7 @@ OBJECT_DECLARE_SIMPLE_TYPE(GenPCIERootPort, GEN_PCIE_ROOT_PORT) (GEN_PCIE_ROOT_PORT_AER_OFFSET + PCI_ERR_SIZEOF) #define GEN_PCIE_ROOT_PORT_MSIX_NR_VECTOR 1 +#define GEN_PCIE_ROOT_DEFAULT_IO_RANGE 4096 struct GenPCIERootPort { /*< private >*/ @@ -75,6 +76,7 @@ static bool gen_rp_test_migrate_msix(void *opaque, int version_id) static void gen_rp_realize(DeviceState *dev, Error **errp) { PCIDevice *d = PCI_DEVICE(dev); + PCIESlot *s = PCIE_SLOT(d); GenPCIERootPort *grp = GEN_PCIE_ROOT_PORT(d); PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(d); Error *local_err = NULL; @@ -85,6 +87,9 @@ static void gen_rp_realize(DeviceState *dev, Error **errp) return; } + if (grp->res_reserve.io == -1 && s->hotplug && !s->native_hotplug) { + grp->res_reserve.io = GEN_PCIE_ROOT_DEFAULT_IO_RANGE; + } int rc = pci_bridge_qemu_reserve_cap_init(d, 0, grp->res_reserve, errp);
Q35 has now ACPI hotplug enabled by default for PCI(e) devices. As opposed to native PCIe hotplug, some guests will not assign IO range to pcie-root-ports not supporting native hotplug, resulting into a regression. Fix it by setting the "reserve-io" hint capability of the pcie-root-ports so the firmware will allocate the IO range instead. Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> --- hw/pci-bridge/gen_pcie_root_port.c | 5 +++++ 1 file changed, 5 insertions(+)