Kconfig: Clean up the PIIX southbridge devices
cover Kconfig: Clean up the PIIX southbridge devices
Hi,
This series is a cleanup for the PIIX/PIIX3/PIIX4 devices.
The isapc machine split isn't clean because while designed on a PIIX3, our model uses ACPI features from the PIIX4.
The i440fx cleanup is left for the the next development cycle.
This series is a prerequisite for the MIPS Malta port to Kconfig.
Please review.
Regards,
Phil.
Philippe Mathieu-Daudé (8):
hw/i2c: The ACPI_SMBUS Kconfig depends on ACPI hw/input: Add 8042 PS/2 Keyboard Controller to Kconfig hw/southbridge: Add the PIIX chipset to Kconfig hw/southbridge: Add the PIIX3 chipset to Kconfig hw/southbridge: Cleanup the PIIX4 chipset in Kconfig hw/southbridge: Add ACPI_PIIX4 to Kconfig
hw/i386: The 'isapc' machine is built around a PIIX3 southbridge hw/i386: Move ACPI_SMBUS out of I440FX (Kconfig)
default-configs/mips-softmmu-common.mak | 1 + hw/acpi/Kconfig | 7 +++++++ hw/acpi/Makefile.objs | 3 ++- hw/i2c/Kconfig | 1 + hw/i386/Kconfig | 9 ++++---- hw/ide/Kconfig | 1 + hw/input/Kconfig | 4 ++++ hw/isa/Kconfig | 28 ++++++++++++++++++++++++- 8 files changed, 47 insertions(+), 7 deletions(-)
Signature · 1 line
2.20.1
PATCH 1/8 hw/i2c: The ACPI_SMBUS Kconfig depends on ACPI
--- hw/i2c/Kconfig | 1 + 1 file changed, 1 insertion(+)
hw/i2c/Kconfig +1 -0
@@ -17,6 +17,7 @@ config VERSATILE_I2C config ACPI_SMBUS bool select I2C + depends on ACPI config BITBANG_I2C bool
Signature · 1 line
2.20.1
Re: [Qemu-devel] [PATCH 1/8] hw/i2c: The ACPI_SMBUS Kconfig depends on ACPI
Quoted 10 lines · Philippe Mathieu-Daudé
Signed-off-by Philippe Mathieu-Daudé <[email protected]>--- hw/i2c/Kconfig | 1 + 1 file changed, 1 insertion(+)hw/i2c/Kconfig +0
-0
hw/i2c/Kconfig config VERSATILE_I2C · line 20
Quoted 4 lines
hw/i2c/Kconfig +1
-0@@ -17,6 +17,7 @@ config VERSATILE_I2C config ACPI_SMBUS bool select I2C + depends on ACPI
Please ignore this patch in this series, I incorrectly reordered it here while rebasing, but it doesn't work without the next series "Kconfig:
Cleanup the ICH9 device". As it there is a circular dependency with PC_ACPI.
Quoted 4 lines
hw/i2c/Kconfig +0
-0config BITBANG_I2C bool
PATCH 2/8 hw/input: Add 8042 PS/2 Keyboard Controller to Kconfig
In reply to Philippe Mathieu-DaudéThe Intel 8042 chipset also includes some functions unrelated to the keyboard, such the A20-Gate.
--- hw/input/Kconfig | 4 ++++ 1 file changed, 4 insertions(+)
hw/input/Kconfig +4 -0
@@ -10,6 +10,10 @@ config PCKBD default y depends on ISA_BUS +config I8042 + bool + select PCKBD + config PL050 bool
Signature · 1 line
2.20.1
Re: [Qemu-devel] [PATCH 2/8] hw/input: Add 8042 PS/2 Keyboard Controller to Kconfig
Quoted 13 lines · Philippe Mathieu-Daudé
The Intel 8042 chipset also includes some functions unrelated to the keyboard, such the A20-Gate.
Signed-off-by Philippe Mathieu-Daudé <[email protected]>--- hw/input/Kconfig | 4 ++++ 1 file changed, 4 insertions(+)hw/input/Kconfig +0
-0
hw/input/Kconfig config PCKBD · line 15
Quoted 6 lines
hw/input/Kconfig +3
-0@@ -10,6 +10,10 @@ config PCKBD default y depends on ISA_BUS +config I8042 + bool + select PCKBD
I fail to see why you need this config switch. There is a "select I8042" in a later patch, but that's it - you never use this switch to decide anything (like including a file in the Makefiles or toggle other switches), so this seems to be just a defunc switch to me. Or do I miss something?
Thomas
Re: [Qemu-devel] [PATCH 2/8] hw/input: Add 8042 PS/2 Keyboard Controller to Kconfig
Quoted 17 lines · Thomas Huth
Quoted 16 lines · Philippe Mathieu-Daudé
The Intel 8042 chipset also includes some functions unrelated to the keyboard, such the A20-Gate.
Signed-off-by Philippe Mathieu-Daudé <[email protected]>--- hw/input/Kconfig | 4 ++++ 1 file changed, 4 insertions(+)hw/input/Kconfig +0
-0@@ -10,6 +10,10 @@ config PCKBD default y depends on ISA_BUS
Quoted 9 lines
hw/input/Kconfig config PCKBD · line 15
Quoted 3 lines
hw/input/Kconfig +3
-0+config I8042 + bool + select PCKBDI fail to see why you need this config switch. There is a "select I8042" in a later patch, but that's it - you never use this switch to decide anything (like including a file in the Makefiles or toggle other switches), so this seems to be just a defunc switch to me. Or do I miss something?
The PIIX3 emulates a 8042 when the USB Legacy Support is enabled.
Oh now I see, I first understood the hw/input/pckbd.c file was modeling the 8042 and other things, but no, it is simply it.
So please disregard this patch, I'll respin renaming PCKBD -> I8042.
Thanks!
Phil.
PATCH 3/8 hw/southbridge: Add the PIIX chipset to Kconfig
In reply to Philippe Mathieu-DaudéThe PIIX (Intel 82371FB) is a bridge between PCI <-> ISA.
It contains:
- IDE interface (PIO and burst transfers)
- one 82c54 timer (and speaker tone output)
- two 82c59 interrupts controllers
- two 8237 DMA controllers
- Power Management (programmable SMI)
- NMI
--- hw/ide/Kconfig | 1 + hw/isa/Kconfig | 13 +++++++++++++ 2 files changed, 14 insertions(+)
hw/ide/Kconfig +1 -0
@@ -17,6 +17,7 @@ config IDE_ISA config IDE_PIIX bool + select IDE_ISA select IDE_PCI select IDE_QDEV
hw/isa/Kconfig +13 -0
@@ -24,6 +24,19 @@ config PC87312 select FDC select IDE_ISA +config PIIX + bool + select PCI_PIIX + select IDE_PIIX + select I8254 + select I8259 + select PCSPK + select I8257 + #select PM_PIIX + #select SMI_PIIX + #select NMI_PIIX + select ISA_BUS + config PIIX4 bool # For historical reasons, SuperIO devices are created in the board
Signature · 1 line
2.20.1
PATCH 4/8 hw/southbridge: Add the PIIX3 chipset to Kconfig
In reply to Philippe Mathieu-DaudéThe PIIX3 (Intel 82371SB) is a bridge between PCI <-> ISA.
It is an exhanced PIIX, thus contains the same features. It also contains:
- separate Master/Slave IDE mode
- compliant to PCI rev 2.1 specifications
- IOAPIC
- USB UHCI (2 ports)
- USB Legacy Support (emulated devices):
- 8042 Keyboard Controller
- A20-Gate
--- hw/isa/Kconfig | 9 +++++++++ 1 file changed, 9 insertions(+)
hw/isa/Kconfig +9 -0
@@ -37,6 +37,15 @@ config PIIX #select NMI_PIIX select ISA_BUS +config PIIX3 + bool + select PIIX + #select PCI_PIIX3 + #select IDE_PIIX3 + select IOAPIC + select USB_UHCI + select I8042 + config PIIX4 bool # For historical reasons, SuperIO devices are created in the board
Signature · 1 line
2.20.1
PATCH 5/8 hw/southbridge: Cleanup the PIIX4 chipset in Kconfig
In reply to Philippe Mathieu-DaudéThe PIIX4 (Intel 82371AB) is a bridge between PCI <-> ISA.
It is an exhanced PIIX3, thus contains the same features. It also contains:
- Power Management (ACPI rev 1.0)
- IDE UDMA/33
- SMBus interface (host and slave)
- RTC
--- hw/isa/Kconfig | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
hw/isa/Kconfig +5 -1
@@ -48,9 +48,13 @@ config PIIX3 config PIIX4 bool + select PIIX3 + #select ACPI_PIIX4 + #select IDE_PIIX4 + select ACPI_SMBUS + select MC146818RTC # For historical reasons, SuperIO devices are created in the board # for PIIX4. - select ISA_BUS config VT82C686 bool
Signature · 1 line
2.20.1
PATCH 6/8 hw/southbridge: Add ACPI_PIIX4 to Kconfig
In reply to Philippe Mathieu-DaudéSince the default-configs/mips-softmmu-common.mak is not yet ported to the Kconfig format, we have to select this variable there (used by the MIPS Malta machine).
--- default-configs/mips-softmmu-common.mak | 1 + hw/acpi/Kconfig | 7 +++++++ hw/acpi/Makefile.objs | 3 ++- hw/i386/Kconfig | 2 ++ hw/isa/Kconfig | 2 +- 5 files changed, 13 insertions(+), 2 deletions(-)
default-configs/mips-softmmu-common.mak +1 -0
@@ -20,6 +20,7 @@ CONFIG_ACPI_X86=y CONFIG_ACPI_MEMORY_HOTPLUG=y CONFIG_ACPI_NVDIMM=y CONFIG_ACPI_CPU_HOTPLUG=y +CONFIG_ACPI_PIIX4=y CONFIG_APM=y CONFIG_I8257=y CONFIG_PIIX4=y
hw/acpi/Kconfig +7 -0
@@ -8,6 +8,13 @@ config ACPI_X86 select ACPI_CPU_HOTPLUG select ACPI_MEMORY_HOTPLUG +config ACPI_PIIX4 + bool + select ACPI + select ACPI_NVDIMM + select ACPI_CPU_HOTPLUG + select ACPI_MEMORY_HOTPLUG + config ACPI_X86_ICH bool select ACPI_X86
hw/acpi/Makefile.objs +2 -1
@@ -1,5 +1,6 @@ ifeq ($(CONFIG_ACPI),y) -common-obj-$(CONFIG_ACPI_X86) += core.o piix4.o pcihp.o +common-obj-$(CONFIG_ACPI_X86) += core.o pcihp.o +common-obj-$(CONFIG_ACPI_PIIX4) += piix4.o common-obj-$(CONFIG_ACPI_X86_ICH) += ich9.o tco.o common-obj-$(CONFIG_ACPI_CPU_HOTPLUG) += cpu_hotplug.o common-obj-$(CONFIG_ACPI_MEMORY_HOTPLUG) += memory_hotplug.o
hw/i386/Kconfig +2 -0
@@ -69,6 +69,8 @@ config ISAPC # FIXME: it is in the same file as i440fx, and does not compile # if separated depends on I440FX + # QEMU isapc weirdness: it uses PIIX3 with ACPI features from PIIX4 + select ACPI_PIIX4 config Q35 bool
hw/isa/Kconfig +1 -1
@@ -49,7 +49,7 @@ config PIIX3 config PIIX4 bool select PIIX3 - #select ACPI_PIIX4 + select ACPI_PIIX4 #select IDE_PIIX4 select ACPI_SMBUS select MC146818RTC
Signature · 1 line
2.20.1
PATCH 7/8 hw/i386: The 'isapc' machine is built around a PIIX3 southbridge
In reply to Philippe Mathieu-Daudé--- hw/i386/Kconfig | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-)
hw/i386/Kconfig +1 -4
@@ -52,8 +52,6 @@ config I440FX select PC_PCI select PC_ACPI select ACPI_SMBUS - select PCI_PIIX - select IDE_PIIX select DIMM select SMBIOS select VMPORT
@@ -62,13 +60,12 @@ config I440FX config ISAPC bool - select ISA_BUS select PC - select IDE_ISA select VGA_ISA # FIXME: it is in the same file as i440fx, and does not compile # if separated depends on I440FX + select PIIX3 # QEMU isapc weirdness: it uses PIIX3 with ACPI features from PIIX4 select ACPI_PIIX4
Signature · 1 line
2.20.1
PATCH 8/8 hw/i386: Move ACPI_SMBUS out of I440FX (Kconfig)
In reply to Philippe Mathieu-DaudéThe SMBus interface is a slow bus not related to the i440fx northbridge, it comes from the PIIX3 southbridge.
--- hw/i386/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
hw/i386/Kconfig +1 -1
@@ -51,7 +51,6 @@ config I440FX bool select PC_PCI select PC_ACPI - select ACPI_SMBUS select DIMM select SMBIOS select VMPORT
@@ -68,6 +67,7 @@ config ISAPC select PIIX3 # QEMU isapc weirdness: it uses PIIX3 with ACPI features from PIIX4 select ACPI_PIIX4 + select ACPI_SMBUS config Q35 bool
Signature · 1 line
2.20.1
Patchew URL: https://patchew.org/QEMU/[email protected]/
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the docker-mingw@fedora build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.
=== TEST SCRIPT BEGIN ===
#!/bin/bash
time make docker-test-mingw@fedora SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===
The full log is available at
http://patchew.org/logs/[email protected]/testing.docker-mingw@fedora/?type=message.
---
Email generated automatically by Patchew [http://patchew.org/].
Please send your feedback to [email protected]