[coreboot] [v2] r4245 - in trunk/coreboot-v2: src/mainboard/rca/rm4100 src/mainboard/thomson/ip1000 targets/rca/rm4100 targets/thomson/ip1000
stepan at coresystems.de
Sat May 2 01:14:36 CEST 2009
Myles Watson wrote:
>>> How will the unconditional inclusion of the vga.o (from the other
>>> thread) affect headless users? It still seems to me that it would be
>>> easy to say
>>> if CONFIG_CONSOLE_VGA
>>> object vga.o
>> Hmm, interesting.
>> I was also thinking about making it a little more user friendly by
>> all the vga options into one. Something like in mainboard Options.lb:
> You have to be careful with #ifdef because our options are always defined.
> You probably meant #if VGA == 1.
>> #ifdef VGA_ENABLE /* or some other global define */
>> default CONFIG_CONSOLE_VGA = 1
>> default CONFIG_PCI_ROM_RUN = 1
>> default CONFIG_PCI_OPTION_ROM_RUN_VM86 = 1
>> default CONFIG_VIDEO_MB = 8
>> default CONFIG_CONSOLE_VGA = 0
>> default CONFIG_PCI_ROM_RUN = 0
>> default CONFIG_PCI_OPTION_ROM_RUN_VM86 = 0
>> default CONFIG_VIDEO_MB = 0
> I would think it would be easier to do something like:
> default CONFIG_PCI_ROM_RUN = CONFIG_CONSOLE_VGA
> default CONFIG_PCI_OPTION_ROM_RUN_VM86 = CONFIG_CONSOLE_VGA
> default CONFIG_VIDEO_MB = CONFIG_CONSOLE_VGA * 8 /* A little hackish. */
Normally, with coreboot and VGA initialization, you don't want
CONFIG_CONSOLE_VGA, as that's just the last few boot messages of table
creation. It's not really good for debugging anymore. In v3 we were
smart and deleted CONFIG_CONSOLE_VGA completely, leaving any VGA
_output_ to the payloads.
I think we should do the same thing in v2.
More information about the coreboot