On 20/2/24 13:55, Thomas Huth wrote:
On 20/02/2024 13.20, BALATON Zoltan wrote:
On Tue, 20 Feb 2024, Philippe Mathieu-Daudé wrote:
On 19/2/24 19:24, BALATON Zoltan wrote:
On Mon, 19 Feb 2024, BALATON Zoltan wrote:
On Mon, 19 Feb 2024, Philippe Mathieu-Daudé wrote:
Expose TYPE_ICH_DMI_PCI_BRIDGE to the new
"hw/pci-bridge/ich_dmi_pci.h" header.
Signed-off-by: Philippe Mathieu-Daudé <phi...@linaro.org>
---
MAINTAINERS | 1 +
include/hw/pci-bridge/ich_dmi_pci.h | 20 ++++++++++++++++++++
include/hw/southbridge/ich9.h | 2 --
hw/pci-bridge/i82801b11.c | 11 ++++-------
4 files changed, 25 insertions(+), 9 deletions(-)
create mode 100644 include/hw/pci-bridge/ich_dmi_pci.h
diff --git a/MAINTAINERS b/MAINTAINERS
index 1b210c5cc1..50507c3dd6 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2609,6 +2609,7 @@ F: hw/acpi/ich9*.c
F: hw/i2c/smbus_ich9.c
F: hw/isa/lpc_ich9.c
F: include/hw/acpi/ich9*.h
+F: include/hw/pci-bridge/ich_dmi_pci.h
F: include/hw/southbridge/ich9.h
PIIX4 South Bridge (i82371AB)
diff --git a/include/hw/pci-bridge/ich_dmi_pci.h
b/include/hw/pci-bridge/ich_dmi_pci.h
new file mode 100644
index 0000000000..7623b32b8e
--- /dev/null
+++ b/include/hw/pci-bridge/ich_dmi_pci.h
@@ -0,0 +1,20 @@
+/*
+ * QEMU ICH4 i82801b11 dmi-to-pci Bridge Emulation
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef HW_PCI_BRIDGE_ICH_D2P_H
+#define HW_PCI_BRIDGE_ICH_D2P_H
+
+#include "qom/object.h"
+#include "hw/pci/pci_bridge.h"
+
+#define TYPE_ICH_DMI_PCI_BRIDGE "i82801b11-bridge"
+OBJECT_DECLARE_SIMPLE_TYPE(I82801b11Bridge, ICH_DMI_PCI_BRIDGE)
+
+struct I82801b11Bridge {
+ PCIBridge parent_obj;
+};
If this class has no fields of its own why does it need its own
state struct defined? You could just set .instance_size =
sizeof(PCIBridge) in the TypeInfo i82801b11_bridge_info below and
delete this struct completely as it's not even used anywhere. One
less needless QOM complication :-) For an example see the empty
via-mc97 device in hw/audio/via-ac97.c.
Then you can put the OBJECT_DECLARE_SIMPLE_TYPE in
hw/pci-bridge/i82801b11.c where this object is defined and the
#define TYPE_ICH_DMI_PCI_BRIDGE in
You don't even need OBJECT_DECLARE_SIMPLE_TYPE if there's no state
struct. But on second look what is this object at all? It's never
instantiated anywhere. Is it used somewhere?
Here my view is we should always define QOM type names in headers
and use them, in particular in the TypeInfo registration. To unify
style and copy/pasting, better use the QOM DECLARE_TYPE macros.
I envision that might help moving toward DSL and have HW modelling
checks done externally, before starting QEMU. But then this is my
view and I dunno about when we'll get that DSL in so I'm OK to
revisit this patch.
The question here is more if we need this object at all because it
wasn't enstantiated before, and after your series it could be
instantiated by a property that's never set. So unless I misunderstood
somthing this whole thing could just be removed as dead code and let
it be re-added later when it's actually implemented following whatever
conventions we'll have then. No need to keep around empty placeholders
that aren't used. Or does it serve any purpose?
This isn't a virtual hardware, and is well specified, I'm trying to
plug all the parts we have so the full chipset can be used to create
a dynamic machine.
It's apparently used by some q35 configs:
$ grep -r i82801b11 docs/
docs/config/q35-emulated.cfg: driver = "i82801b11-bridge"
Thomas
