On 10.10.2008 09:25, Corey Osgood wrote: > On Fri, Oct 10, 2008 at 12:46 AM, Corey Osgood <[EMAIL PROTECTED]>wrote: > > >> On Thu, Oct 9, 2008 at 9:57 PM, Corey Osgood <[EMAIL PROTECTED]>wrote: >> >> >>> On Thu, Oct 9, 2008 at 7:56 PM, ron minnich <[EMAIL PROTECTED]> wrote: >>> >>> >>>> exciting! I am looking forward to this one! >>>> >>> >>> Problems already: >>> >>> [EMAIL PROTECTED]:~/coreboot/coreboot-v3$ make >>> CP build/config.h >>> GEN build/build.h >>> CC build/mainboard/jetway/j7f2/stage1.o >>> DTC build/statictree.h >>> DTC mainboard/jetway/j7f2/dts (dts->lbh) >>> CC build/superio/fintek/f71805f/stage1.o >>> CC build/arch/x86/via/stage0.o >>> AS build/arch/x86/via/stage0.o >>> /home/corey/coreboot/coreboot-v3/arch/x86/via/stage0.S: Assembler >>> messages: >>> /home/corey/coreboot/coreboot-v3/arch/x86/via/stage0.S:37: Error: can't >>> handle non absolute segment in `ljmp' >>> >>> >> No progress on this. Makes no sense to me, geode and i586 car use the same >> piece of asm without complaint >> >> > > Sorry for the mailbox flooding, I'm slowly learning to step back and look at > things before jumping to conclusions. Geode and i586 also have ROM_CODE_SEG > defined, this didn't, that's why the jump was going to nowhere. Do those > defines belong in this file, somewhere else, or in Kconfig? Are the i586 > values the same as they should be for c7? > > I'm now getting hung up because XIP_ROM_[SIZE,BASE] isn't defined. Do we > still need XIP and those defines? Are the value limitations the same as they > were in v2? >
To be honest, I'm debating whether we should make the XIP stuff optional for all arches. While it may cause speedups, it will also introduce a noticeable delay in the VIA CAR implementation before we enter C code and that can be confusing. Regards, Carl-Daniel -- http://www.hailfinger.org/ -- coreboot mailing list: coreboot@coreboot.org http://www.coreboot.org/mailman/listinfo/coreboot
