Re: boot(8) for bebox
KIYOHARA Takashi <[email protected]> Thu, 03 Apr 2008 18:01:28 +0900 (JST)
| Newsgroups | gmane.os.netbsd.ports.bebox,gmane.os.netbsd.ports.powerpc |
|---|---|
| Message-ID | <[email protected]> |
Hi! Tim, From: Tim Rightnour <[email protected]> Date: Thu, 03 Apr 2008 00:56:16 -0700 (MST) > On 03-Apr-2008 KIYOHARA Takashi wrote: > > I will split boot(8) of bebox into boot and boot_com0 now, I am trying. > > > > However, I had forgotten CONS_BE existed in bebox. It is necessary to > > depend on the VGA card in bebox, and to be booted by either of CONS_VGA > > or CONS_BE. > > > > I have not understood yet, and when BootROM of BeBox displays on the > > screen maybe as "Be", I think that it boots with CONS_BE. > > What is cons_be? Is that more like a framebuffer console? I got rid of the > prep framebuffer console, as it was in really bad shape (looked like it was > copied from bebox). What I really want to see some day in the future is libsa > containing enough rasops code so that a proper genfb driver can be written > inside a bootloader, and we use that instead. > > > Is boot_be necessary for us? > > Or, do boot support both CONS_VGA and CONS_BE, and do it correspond with > > Vendor/Product ID of VGA? > > If you can somehow boot the machine and generate text output on the console, > I'd just get rid of it. The only reason to keep it would be if there was no > other way to get console output. boot build with -DCONS_BE, video_{init,put} builded in video.c. It supports a another frame buffer from vga_{init,put} in vga.c. It has been described to sys/arch/bebox/stand/README. Bebox has three PCI empty slots, and the VGA card has stuck in the our one. I think that various PCI VGA cards had stuck in BeBox though I do not understand. And, it will support either. For instance, if we rid video.c, my Trio64v+ cannot be used on bebox. Thanks, -- kiyohara