From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:51066) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UOVlD-0005s9-96 for qemu-devel@nongnu.org; Sat, 06 Apr 2013 12:14:42 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UOVl9-0006U6-QH for qemu-devel@nongnu.org; Sat, 06 Apr 2013 12:14:35 -0400 Received: from 1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.d.1.0.0.b.8.0.1.0.0.2.ip6.arpa ([2001:8b0:1d0::1]:33622 helo=mnementh.archaic.org.uk) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UOVl9-0006U1-JK for qemu-devel@nongnu.org; Sat, 06 Apr 2013 12:14:31 -0400 From: Peter Maydell Date: Sat, 6 Apr 2013 16:44:30 +0100 Message-Id: <1365263076-16909-6-git-send-email-peter.maydell@linaro.org> In-Reply-To: <1365263076-16909-1-git-send-email-peter.maydell@linaro.org> References: <1365263076-16909-1-git-send-email-peter.maydell@linaro.org> Subject: [Qemu-devel] [PATCH v4 05/11] versatile_pci: Use separate PCI I/O space rather than system I/O space List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Arnd Bergmann , "Michael S. Tsirkin" , patches@linaro.org, Will Deacon , Paul Brook , =?UTF-8?q?Andreas=20F=C3=A4rber?= Rather than overloading the system I/O space (which doesn't even make any sense on ARM) for PCI I/O, create an memory region in the PCI controller and use that to represent the I/O space. Signed-off-by: Peter Maydell Acked-by: Paul Brook --- hw/versatile_pci.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/hw/versatile_pci.c b/hw/versatile_pci.c index 9e0ece0..ce5bdf2 100644 --- a/hw/versatile_pci.c +++ b/hw/versatile_pci.c @@ -19,7 +19,8 @@ typedef struct { qemu_irq irq[4]; MemoryRegion mem_config; MemoryRegion mem_config2; - MemoryRegion isa; + MemoryRegion pci_io_space; + MemoryRegion pci_io_window; PCIBus pci_bus; PCIDevice pci_dev; @@ -77,8 +78,10 @@ static void pci_vpb_init(Object *obj) PCIHostState *h = PCI_HOST_BRIDGE(obj); PCIVPBState *s = PCI_VPB(obj); + memory_region_init(&s->pci_io_space, "pci_io", 1ULL << 32); + pci_bus_new_inplace(&s->pci_bus, DEVICE(obj), "pci", - get_system_memory(), get_system_io(), + get_system_memory(), &s->pci_io_space, PCI_DEVFN(11, 0), TYPE_PCI_BUS); h->bus = &s->pci_bus; @@ -111,8 +114,14 @@ static void pci_vpb_realize(DeviceState *dev, Error **errp) memory_region_init_io(&s->mem_config2, &pci_vpb_config_ops, &s->pci_bus, "pci-vpb-config", 0x1000000); sysbus_init_mmio(sbd, &s->mem_config2); - isa_mmio_setup(&s->isa, 0x0100000); - sysbus_init_mmio(sbd, &s->isa); + + /* The window into I/O space is always into a fixed base address; + * its size is the same for both realview and versatile. + */ + memory_region_init_alias(&s->pci_io_window, "pci-vbp-io-window", + &s->pci_io_space, 0, 0x100000); + + sysbus_init_mmio(sbd, &s->pci_io_space); /* TODO Remove once realize propagates to child devices. */ object_property_set_bool(OBJECT(&s->pci_dev), true, "realized", errp); -- 1.7.9.5