coreboot August 2008

coreboot@coreboot.org

64 participants
593 discussions

[PATCH] v3: missing 640k-1024k mem hole for i440bxemulation
by Carl-Daniel Hailfinger 22 Aug '08

22 Aug '08

i440bxemulation is missing the 640k-1024k mem hole. geodelx has the hole. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: LinuxBIOSv3-stuff/northbridge/intel/i440bxemulation/i440bx.c =================================================================== --- LinuxBIOSv3-stuff/northbridge/intel/i440bxemulation/i440bx.c (Revision 644) +++ LinuxBIOSv3-stuff/northbridge/intel/i440bxemulation/i440bx.c (Arbeitskopie) @@ -59,7 +59,10 @@ mc_dev = dev->link[0].children; if (mc_dev) { idx = 10; - ram_resource(dev, idx++, 0, tolmk); + /* 0 .. 640 KB */ + ram_resource(dev, idx++, 0, 640); + /* 1 MB .. (Systop - 1 MB) (in KB) */ + ram_resource(dev, idx++, 1024, tolmk - 1024); } phase4_assign_resources(&dev->link[0]); } -- http://www.hailfinger.org/

4 6

r802 - coreboot-v3/northbridge/intel/i440bxemulation
by svn＠coreboot.org 22 Aug '08

22 Aug '08

Author: hailfinger Date: 2008-08-22 13:29:30 +0200 (Fri, 22 Aug 2008) New Revision: 802 Modified: coreboot-v3/northbridge/intel/i440bxemulation/i440bx.c Log: i440bxemulation is missing the hole between 640k and 768k for VGA (0xA0000-0xAFFFF) and text mode (0xB8000-0xBFFFF). Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Acked-by: Ronald G. Minnich <rminnich(a)gmail.com> Modified: coreboot-v3/northbridge/intel/i440bxemulation/i440bx.c =================================================================== --- coreboot-v3/northbridge/intel/i440bxemulation/i440bx.c 2008-08-22 11:10:04 UTC (rev 801) +++ coreboot-v3/northbridge/intel/i440bxemulation/i440bx.c 2008-08-22 11:29:30 UTC (rev 802) @@ -61,8 +61,12 @@ mc_dev = dev->link[0].children; if (mc_dev) { idx = 10; -#warning FIXME: We have no memory hole between 640 and 768 kB - ram_resource(dev, idx++, 0, tolmk); + /* 0 .. 640 kB */ + ram_resource(dev, idx++, 0, 640); + /* Hole for VGA (0xA0000-0xAFFFF) graphics and text mode + * graphics (0xB8000-0xBFFFF) */ + /* 768 kB .. Systop (in KB) */ + ram_resource(dev, idx++, 768, tolmk - 768); } phase4_assign_resources(&dev->link[0]); }

1 0

[PATCH] v3: improve PCI device doxygen comments and printks
by Carl-Daniel Hailfinger 22 Aug '08

22 Aug '08

Improve/fix PCI device doxygen comments and printks. As a bonus, the boot messages are more understandable now. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: corebootv3-pci_device_better_prints_comments/device/pci_device.c =================================================================== --- corebootv3-pci_device_better_prints_comments/device/pci_device.c (Revision 759) +++ corebootv3-pci_device_better_prints_comments/device/pci_device.c (Arbeitskopie) @@ -929,12 +929,15 @@ /** * Scan a PCI bus. * - * Determine the existence of a given PCI device. + * Determine the existence of a given PCI device. Allocate a new struct device + * if dev==NULL was passed in and the device exists. * + * @param dev Pointer to the device structure if it already exists or NULL * @param bus Pointer to the bus structure. * @param devfn A device/function number. * @return The device structure for the device (if found) - * or the NULL if no device is found. + * or NULL if no device is found and dev==NULL was passed in + * or the device structure with enabled=0 if no device is found. */ struct device *pci_probe_dev(struct device *dev, struct bus *bus, unsigned int devfn) @@ -990,8 +993,8 @@ if ((id == 0xffffffff) || (id == 0x00000000) || (id == 0x0000ffff) || (id == 0xffff0000)) { if (dev->enabled) { - printk(BIOS_INFO, - "Disabling static device: %s\n", + printk(BIOS_INFO, "PCI: Static device not " + "found, setting enabled=0: %s\n", dev_path(dev)); dev->enabled = 0; } -- http://www.hailfinger.org/

3 9

r801 - coreboot-v3/device
by svn＠coreboot.org 22 Aug '08

22 Aug '08

Author: hailfinger Date: 2008-08-22 13:10:04 +0200 (Fri, 22 Aug 2008) New Revision: 801 Modified: coreboot-v3/device/pci_device.c Log: Improve/fix PCI device doxygen comments. Thanks to Ron for explaining this. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Acked-by: Ronald G. Minnich <rminnich(a)gmail.com> Modified: coreboot-v3/device/pci_device.c =================================================================== --- coreboot-v3/device/pci_device.c 2008-08-22 10:57:34 UTC (rev 800) +++ coreboot-v3/device/pci_device.c 2008-08-22 11:10:04 UTC (rev 801) @@ -950,12 +950,19 @@ /** * Scan a PCI bus. * - * Determine the existence of a given PCI device. + * Determine the existence of a given PCI device. Allocate a new struct device + * if dev==NULL was passed in and the device exists in hardware. * + * @param dev Pointer to the device structure if it already is in the device + * tree, i.e. was specified in the dts. It may not exist on hardware, + * however. Looking for hardware not yet in the device tree has this + * parameter as NULL. * @param bus Pointer to the bus structure. * @param devfn A device/function number. - * @return The device structure for the device (if found) - * or the NULL if no device is found. + * @return The device structure for the device if it exists in hardware + * or the passed in device structure with enabled=0 if the device + * does not exist in hardware and only in the tree + * or NULL if no device is found and dev==NULL was passed in. */ struct device *pci_probe_dev(struct device *dev, struct bus *bus, unsigned int devfn)

1 0

[PATCH] v3: Add Kconfig for CAR testing code
by Carl-Daniel Hailfinger 22 Aug '08

22 Aug '08

Add Kconfig for CAR testing code. The code is already in the tree, but we never used it. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: corebootv3-cartest/arch/x86/Kconfig =================================================================== --- corebootv3-cartest/arch/x86/Kconfig (Revision 796) +++ corebootv3-cartest/arch/x86/Kconfig (Arbeitskopie) @@ -231,3 +231,13 @@ depends IOAPIC help This is entirely mainboard dependent. 0x10 is a *typical* setting but not always a good setting. + +menu "Debugging" + +config CARTEST + bool "Test CAR area" + default n + help + Test the CAR area after it has been set up. + +endmenu -- http://www.hailfinger.org/

2 2

r800 - coreboot-v3/arch/x86
by svn＠coreboot.org 22 Aug '08

22 Aug '08

Author: hailfinger Date: 2008-08-22 12:57:34 +0200 (Fri, 22 Aug 2008) New Revision: 800 Modified: coreboot-v3/arch/x86/Kconfig Log: Add Kconfig for CAR testing code. The code is already in the tree, but we never used it. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Acked-by: Ronald G. Minnich <rminnich(a)gmail.com> Modified: coreboot-v3/arch/x86/Kconfig =================================================================== --- coreboot-v3/arch/x86/Kconfig 2008-08-22 10:56:36 UTC (rev 799) +++ coreboot-v3/arch/x86/Kconfig 2008-08-22 10:57:34 UTC (rev 800) @@ -231,3 +231,13 @@ depends IOAPIC help This is entirely mainboard dependent. 0x10 is a *typical* setting but not always a good setting. + +menu "Debugging" + +config CARTEST + bool "Test CAR area" + default n + help + Test the CAR area after it has been set up. + +endmenu

1 0

[PATCH] v3: smbus_delay calls udelay
by Carl-Daniel Hailfinger 22 Aug '08

22 Aug '08

smbus_delay() performs its own inb(0x80). We can use the generic udelay() instead which does the same, and achieve better abstraction. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: corebootv3-stuff3/southbridge/amd/cs5536/smbus_initram.c =================================================================== --- corebootv3-stuff3/southbridge/amd/cs5536/smbus_initram.c (Revision 798) +++ corebootv3-stuff3/southbridge/amd/cs5536/smbus_initram.c (Arbeitskopie) @@ -69,7 +69,7 @@ */ static void smbus_delay(void) { - /* inb(0x80); */ + /* udelay(1); */ } /** Index: corebootv3-stuff3/southbridge/nvidia/mcp55/mcp55_smbus.h =================================================================== --- corebootv3-stuff3/southbridge/nvidia/mcp55/mcp55_smbus.h (Revision 798) +++ corebootv3-stuff3/southbridge/nvidia/mcp55/mcp55_smbus.h (Arbeitskopie) @@ -39,8 +39,7 @@ static inline void smbus_delay(void) { - /* let's hope this is not optimized out */ - (void) inb(0x80); + udelay(1); } int do_smbus_recv_byte(u16 smbus_io_base, u8 device); -- http://www.hailfinger.org/

2 2

r799 - in coreboot-v3/southbridge: amd/cs5536 nvidia/mcp55
by svn＠coreboot.org 22 Aug '08

22 Aug '08

Author: hailfinger Date: 2008-08-22 12:56:36 +0200 (Fri, 22 Aug 2008) New Revision: 799 Modified: coreboot-v3/southbridge/amd/cs5536/smbus_initram.c coreboot-v3/southbridge/nvidia/mcp55/mcp55_smbus.h Log: smbus_delay() performs its own inb(0x80). We can use the generic udelay() instead which does the same, and achieve better abstraction. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Acked-by: Ronald G. Minnich <rminnich(a)gmail.com> Modified: coreboot-v3/southbridge/amd/cs5536/smbus_initram.c =================================================================== --- coreboot-v3/southbridge/amd/cs5536/smbus_initram.c 2008-08-22 01:22:21 UTC (rev 798) +++ coreboot-v3/southbridge/amd/cs5536/smbus_initram.c 2008-08-22 10:56:36 UTC (rev 799) @@ -69,7 +69,7 @@ */ static void smbus_delay(void) { - /* inb(0x80); */ + /* udelay(1); */ } /** Modified: coreboot-v3/southbridge/nvidia/mcp55/mcp55_smbus.h =================================================================== --- coreboot-v3/southbridge/nvidia/mcp55/mcp55_smbus.h 2008-08-22 01:22:21 UTC (rev 798) +++ coreboot-v3/southbridge/nvidia/mcp55/mcp55_smbus.h 2008-08-22 10:56:36 UTC (rev 799) @@ -39,8 +39,7 @@ static inline void smbus_delay(void) { - /* let's hope this is not optimized out */ - (void) inb(0x80); + udelay(1); } int do_smbus_recv_byte(u16 smbus_io_base, u8 device);

1 0

[PATCH] v3: use inb instead of outb for delays
by Carl-Daniel Hailfinger 22 Aug '08

22 Aug '08

Use inb instead of outb for delays. v2 has already been converted in r3509 which was Signed-off-by: Stefan Reinauer <stepan(a)coresystems.de> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> I'm tempted to do the same here, but with signoff and ack switched. Index: corebootv3-stuff3/arch/x86/udelay_io.c =================================================================== --- corebootv3-stuff3/arch/x86/udelay_io.c (Revision 798) +++ corebootv3-stuff3/arch/x86/udelay_io.c (Arbeitskopie) @@ -23,6 +23,7 @@ void udelay(unsigned int usecs) { unsigned int i; + /* let's hope this is not optimized out */ for (i = 0; i < usecs; i++) - outb(i & 0xff, 0x80); + (void) inb(0x80); } -- http://www.hailfinger.org/

1 0

[PATCH] v3: add objdump and readelf to xcompile
by Carl-Daniel Hailfinger 22 Aug '08

22 Aug '08

Add objdump and readelf to xcompile and use the results. This kills a few hardcodes introduced with my checker. I'd appreciate tests with non-Linux machines. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: corebootv3-check_illegal_global_vars/Makefile =================================================================== --- corebootv3-check_illegal_global_vars/Makefile (Revision 794) +++ corebootv3-check_illegal_global_vars/Makefile (Arbeitskopie) @@ -96,6 +96,8 @@ LD := $(LD_$(ARCH)) NM := $(NM_$(ARCH)) OBJCOPY := $(OBJCOPY_$(ARCH)) +OBJDUMP := $(OBJDUMP_$(ARCH)) +READELF := $(READELF_$(ARCH)) AR := $(AR_$(ARCH)) CFLAGS += $(CFLAGS_$(ARCH)) Index: corebootv3-check_illegal_global_vars/util/xcompile/xcompile =================================================================== --- corebootv3-check_illegal_global_vars/util/xcompile/xcompile (Revision 794) +++ corebootv3-check_illegal_global_vars/util/xcompile/xcompile (Arbeitskopie) @@ -63,6 +63,8 @@ searchgnu ld >/dev/null && echo "LD_x86 := $(searchgnu ld) -b elf32-i386 -melf_i386" searchgnu nm >/dev/null && echo "NM_x86 := $(searchgnu nm)" searchgnu objcopy >/dev/null && echo "OBJCOPY_x86 := $(searchgnu objcopy)" + searchgnu objdump >/dev/null && echo "OBJDUMP_x86 := $(searchgnu objdump)" + searchgnu readelf >/dev/null && echo "READELF_x86 := $(searchgnu readelf)" ;; "x86") if [ "$OS" != "Darwin" ]; then @@ -74,6 +76,8 @@ searchgnu ld >/dev/null && echo "LD_x86 := $(searchgnu ld)" searchgnu nm >/dev/null && echo "NM_x86 := $(searchgnu nm)" searchgnu objcopy >/dev/null && echo "OBJCOPY_x86 := $(searchgnu objcopy)" + searchgnu objdump >/dev/null && echo "OBJDUMP_x86 := $(searchgnu objdump)" + searchgnu readelf >/dev/null && echo "READELF_x86 := $(searchgnu readelf)" else CC=`find /opt/local/bin/ -name "i386-elf-gcc-*"` if [ "$CC" == "" ]; then @@ -86,6 +90,8 @@ echo "LD_x86 := i386-elf-ld" echo "NM_x86 := i386-elf-nm" echo "OBJCOPY_x86 := i386-elf-objcopy" + echo "OBJDUMP_x86 := i386-elf-objdump" + echo "READELF_x86 := i386-elf-readelf" echo "INTLCFLAGS := -I/opt/local/include" echo "INTLLIBS := -L/opt/local/lib -lintl" fi @@ -99,6 +105,8 @@ echo "LD_x86 := i386-linux-ld" echo "NM_x86 := i386-linux-nm" echo "OBJCOPY_x86 := i386-linux-objcopy" + echo "OBJDUMP_x86 := i386-linux-objdump" + echo "READELF_x86 := i386-elf-readelf" ;; esac Index: corebootv3-check_illegal_global_vars/arch/x86/Makefile =================================================================== --- corebootv3-check_illegal_global_vars/arch/x86/Makefile (Revision 794) +++ corebootv3-check_illegal_global_vars/arch/x86/Makefile (Arbeitskopie) @@ -145,7 +145,7 @@ $(Q)# .data and .bss must be empty because they aren't handled $(Q)printf " CHECK stage0 (non-empty writable/allocatable sections)\n" - $(Q)./util/sectionchecker/sectionchecker objdump readelf $(STAGE0_OBJ) + $(Q)./util/sectionchecker/sectionchecker $(OBJDUMP) $(READELF) $(STAGE0_OBJ) $(Q)# Note: we invoke gcc (instead of ld directly) here, as we hit $(Q)# strange problems in the past. It seems that only gcc knows how @@ -271,7 +271,7 @@ $(Q)# .data and .bss must be empty because they aren't handled $(Q)printf " CHECK initram (non-empty writable/allocatable sections)\n" - $(Q)./util/sectionchecker/sectionchecker objdump readelf $(obj)/coreboot.initram_partiallylinked.o + $(Q)./util/sectionchecker/sectionchecker $(OBJDUMP) $(READELF) $(obj)/coreboot.initram_partiallylinked.o $(Q)printf " WRAP $(subst $(shell pwd)/,,$(@)) (PIC->non-PIC)\n" $(Q)$(NM) --undefined-only $(obj)/coreboot.initram_partiallylinked.o |\ -- http://www.hailfinger.org/

1 1

← Newer
1
...
16
17
18
19
20
21
22
...
60
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

coreboot August 2008