Hi,
as preparation for the kconfig merge, here are some changes from the kconfig tree that are unrelated to kconfig itself.
20090805-1-ttys0-rename Takes care of the different CONFIG_* names of the variable containing the serial I/O port
20090805-2-config-prefixes Some more CONFIG_* prefixes that were missing
20090805-3-empty-files These changes exist mostly to make diff happy, which ignores empty files (or rather: Make Patrick happy when he's using diff). Maybe the right way would be to delete those files.
Signed-off-by: Patrick Georgi patrick.georgi@coresystems.de
Index: coreboot-v2/src/arch/i386/lib/printk_init.c =================================================================== --- coreboot-v2.orig/src/arch/i386/lib/printk_init.c +++ coreboot-v2/src/arch/i386/lib/printk_init.c @@ -8,6 +8,10 @@ #include <stdarg.h> #include <console/loglevel.h>
+#if !defined(CONFIG_TTYS0_BASE) +#define CONFIG_TTYS0_BASE CONFIG_SERIAL_IOBASE +#endif + /* printk's without a loglevel use this.. */ #define DEFAULT_MESSAGE_LOGLEVEL 4 /* BIOS_WARNING */
Index: coreboot-v2/src/include/uart8250.h =================================================================== --- coreboot-v2.orig/src/include/uart8250.h +++ coreboot-v2/src/include/uart8250.h @@ -1,6 +1,10 @@ #ifndef UART8250_H #define UART8250_H
+#if !defined(CONFIG_TTYS0_BASE) +#define CONFIG_TTYS0_BASE CONFIG_SERIAL_IOBASE +#endif + struct uart8250 { unsigned int baud; /* Do I need an lcs parameter here? */
Index: coreboot-v2/src/southbridge/intel/i82801gx/i82801gx_lpc.c =================================================================== --- coreboot-v2.orig/src/southbridge/intel/i82801gx/i82801gx_lpc.c +++ coreboot-v2/src/southbridge/intel/i82801gx/i82801gx_lpc.c @@ -273,7 +273,7 @@ static void i82801gx_power_options(devic /* Set up power management block and determine sleep mode */ pmbase = pci_read_config16(dev, 0x40) & 0xfffe; reg32 = inl(pmbase + 0x04); // PM1_CNT -#if HAVE_ACPI_RESUME +#if CONFIG_HAVE_ACPI_RESUME acpi_slp_type = (((reg32 >> 10) & 7) == 5) ? 3 : 0; printk_debug("PM1_CNT: 0x%08x --> acpi_sleep_type: %x\n", reg32, acpi_slp_type); Index: coreboot-v2/src/southbridge/intel/i82801gx/i82801gx_power.h =================================================================== --- coreboot-v2.orig/src/southbridge/intel/i82801gx/i82801gx_power.h +++ coreboot-v2/src/southbridge/intel/i82801gx/i82801gx_power.h @@ -23,7 +23,7 @@ #define MAINBOARD_POWER_ON 1 #define MAINBOARD_POWER_KEEP 2
-#ifndef MAINBOARD_POWER_ON_AFTER_FAIL -#define MAINBOARD_POWER_ON_AFTER_FAIL MAINBOARD_POWER_ON +#ifndef CONFIG_MAINBOARD_POWER_ON_AFTER_POWER_FAIL +#define CONFIG_MAINBOARD_POWER_ON_AFTER_POWER_FAIL MAINBOARD_POWER_ON #endif
Index: coreboot-v2/src/southbridge/intel/i82801gx/i82801gx_smihandler.c =================================================================== --- coreboot-v2.orig/src/southbridge/intel/i82801gx/i82801gx_smihandler.c +++ coreboot-v2/src/southbridge/intel/i82801gx/i82801gx_smihandler.c @@ -300,7 +300,7 @@ static void southbridge_smi_sleep(unsign * CMOS or even better from GNVS. Right now it's hard * coded at compile time. */ - u8 s5pwr = MAINBOARD_POWER_ON_AFTER_FAIL; + u8 s5pwr = CONFIG_MAINBOARD_POWER_ON_AFTER_POWER_FAIL;
/* First, disable further SMIs */ reg8 = inb(pmbase + SMI_EN);
Index: coreboot-v2/src/cpu/x86/32bit/entry32.lds =================================================================== --- /dev/null +++ coreboot-v2/src/cpu/x86/32bit/entry32.lds @@ -0,0 +1 @@ +# Index: coreboot-v2/src/cpu/x86/fpu/Config.lb =================================================================== --- /dev/null +++ coreboot-v2/src/cpu/x86/fpu/Config.lb @@ -0,0 +1 @@ +# Index: coreboot-v2/src/cpu/x86/mmx/Config.lb =================================================================== --- /dev/null +++ coreboot-v2/src/cpu/x86/mmx/Config.lb @@ -0,0 +1 @@ +# Index: coreboot-v2/src/cpu/x86/sse/Config.lb =================================================================== --- /dev/null +++ coreboot-v2/src/cpu/x86/sse/Config.lb @@ -0,0 +1 @@ +# Index: coreboot-v2/src/drivers/generic/generic/Config.lb =================================================================== --- /dev/null +++ coreboot-v2/src/drivers/generic/generic/Config.lb @@ -0,0 +1 @@ +# Index: coreboot-v2/src/northbridge/intel/i855pm/i855pm.h =================================================================== --- /dev/null +++ coreboot-v2/src/northbridge/intel/i855pm/i855pm.h @@ -0,0 +1 @@ +#
Hi Patrick,
On 05.08.2009 13:04, Patrick Georgi wrote:
as preparation for the kconfig merge, here are some changes from the kconfig tree that are unrelated to kconfig itself.
20090805-1-ttys0-rename Takes care of the different CONFIG_* names of the variable containing the serial I/O port
There should only be one config variable for this. Wouldn't a global search/replace be able to standardize on either CONFIG_TTYS0_BASE or CONFIG_SERIAL_IOBASE?
20090805-2-config-prefixes Some more CONFIG_* prefixes that were missing
I don't particularly like the CONFIG_MAINBOARD_POWER_ON_AFTER_POWER_FAIL #define, but since that code already looks like this and you only added a prefix, it is Acked-by: Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net
20090805-3-empty-files These changes exist mostly to make diff happy, which ignores empty files (or rather: Make Patrick happy when he's using diff). Maybe the right way would be to delete those files.
If deleting these files doesn't break abuild, please delete them. For the deletion, you can use my ack.
Signed-off-by: Patrick Georgi patrick.georgi@coresystems.de
Regards, Carl-Daniel
Carl-Daniel Hailfinger schrieb:
Hi Patrick,
On 05.08.2009 13:04, Patrick Georgi wrote:
as preparation for the kconfig merge, here are some changes from the kconfig tree that are unrelated to kconfig itself.
20090805-1-ttys0-rename Takes care of the different CONFIG_* names of the variable containing the serial I/O port
There should only be one config variable for this. Wouldn't a global search/replace be able to standardize on either CONFIG_TTYS0_BASE or CONFIG_SERIAL_IOBASE?
I _think_ the CONFIG_SERIAL_IOBASE was introduced in the kconfig stuff to be more similar to v3. Once we decide on one or the other, I can replace the respective occurrences. I don't care much and would go for CONFIG_TTYS0_BASE as that is what exists in the codebase. I'll prepare a patch.
20090805-2-config-prefixes Some more CONFIG_* prefixes that were missing
I don't particularly like the CONFIG_MAINBOARD_POWER_ON_AFTER_POWER_FAIL #define, but since that code already looks like this and you only added a prefix, it is Acked-by: Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net
Maybe we can get rid of this eventually, but for now this gets in. One battle at a time. r4497
20090805-3-empty-files These changes exist mostly to make diff happy, which ignores empty files (or rather: Make Patrick happy when he's using diff). Maybe the right way would be to delete those files.
If deleting these files doesn't break abuild, please delete them. For the deletion, you can use my ack.
It will require some more work, as these empty files are still included at some places I think. I'll keep this open for discussion.
Thanks, Patrick
-
Carl-Daniel Hailfinger -
Patrick Georgi