| Message ID | 1362157061-438-1-git-send-email-peter.maydell@linaro.org |
|---|---|
| State | New |
| Headers | show |
Am 01.03.2013 17:57, schrieb Peter Maydell: > The address_space field of PCIHostState was only ever written, never used. > Drop it completely. > > Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Andreas Färber <afaerber@suse.de> Andreas
On 1 March 2013 18:33, Andreas Färber <afaerber@suse.de> wrote: > Am 01.03.2013 17:57, schrieb Peter Maydell: >> The address_space field of PCIHostState was only ever written, never used. >> Drop it completely. >> >> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> > > Reviewed-by: Andreas Färber <afaerber@suse.de> Incidentally I suspect you could if you liked get rid of all the uses of PCIHostState in prep.c (and the #include of pci_host.h), since it's now only using &pcihost->busdev as an opaque way of saying SYS_BUS_DEVICE(dev)). -- PMM
Am 01.03.2013 19:50, schrieb Peter Maydell: > On 1 March 2013 18:33, Andreas Färber <afaerber@suse.de> wrote: >> Am 01.03.2013 17:57, schrieb Peter Maydell: >>> The address_space field of PCIHostState was only ever written, never used. >>> Drop it completely. >>> >>> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> >> >> Reviewed-by: Andreas Färber <afaerber@suse.de> > > Incidentally I suspect you could if you liked get rid of all > the uses of PCIHostState in prep.c (and the #include of pci_host.h), > since it's now only using &pcihost->busdev as an opaque way of > saying SYS_BUS_DEVICE(dev)). I believe I do have PCI_DEVICE() and ISA_BUS() changes in my tree. :) Thanks, Andreas
On Fri, Mar 01, 2013 at 04:57:41PM +0000, Peter Maydell wrote: > The address_space field of PCIHostState was only ever written, never used. > Drop it completely. > > Signed-off-by: Peter Maydell <peter.maydell@linaro.org> > --- > Noticed this when looking at converting versatile_pci to be a subclass > of pci_host; it's nice to be able to drop the ugly fishing around in > the object struct that the two users here were doing... > > hw/pci/pci_host.h | 1 - > hw/piix_pci.c | 1 - > hw/ppc/prep.c | 1 - > 3 files changed, 3 deletions(-) Thanks, applied to the trivial patches tree: https://github.com/stefanha/qemu/commits/trivial-patches Stefan
diff --git a/hw/pci/pci_host.h b/hw/pci/pci_host.h index 1845d4d..236cd0f 100644 --- a/hw/pci/pci_host.h +++ b/hw/pci/pci_host.h @@ -40,7 +40,6 @@ struct PCIHostState { MemoryRegion conf_mem; MemoryRegion data_mem; MemoryRegion mmcfg; - MemoryRegion *address_space; uint32_t config_reg; PCIBus *bus; }; diff --git a/hw/piix_pci.c b/hw/piix_pci.c index 6c77e49..9246983 100644 --- a/hw/piix_pci.c +++ b/hw/piix_pci.c @@ -244,7 +244,6 @@ static PCIBus *i440fx_common_init(const char *device_name, dev = qdev_create(NULL, "i440FX-pcihost"); s = PCI_HOST_BRIDGE(dev); - s->address_space = address_space_mem; b = pci_bus_new(dev, NULL, pci_address_space, address_space_io, 0); s->bus = b; diff --git a/hw/ppc/prep.c b/hw/ppc/prep.c index e06dded..2920911 100644 --- a/hw/ppc/prep.c +++ b/hw/ppc/prep.c @@ -567,7 +567,6 @@ static void ppc_prep_init(QEMUMachineInitArgs *args) dev = qdev_create(NULL, "raven-pcihost"); pcihost = PCI_HOST_BRIDGE(dev); - pcihost->address_space = get_system_memory(); object_property_add_child(qdev_get_machine(), "raven", OBJECT(dev), NULL); qdev_init_nofail(dev); pci_bus = (PCIBus *)qdev_get_child_bus(dev, "pci.0");
The address_space field of PCIHostState was only ever written, never used. Drop it completely. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- Noticed this when looking at converting versatile_pci to be a subclass of pci_host; it's nice to be able to drop the ugly fishing around in the object struct that the two users here were doing... hw/pci/pci_host.h | 1 - hw/piix_pci.c | 1 - hw/ppc/prep.c | 1 - 3 files changed, 3 deletions(-)