> > + if (!have_vga) {
> > + // no VGA, try fallback to display
> > + dprintf(1, "no vga, try display\n");
> > + foreachpci(pci) {
> > + if (!is_pci_display_other(pci))
> > + continue;
> > + vgahook_setup(pci);
> > + init_pcirom(pci, 1, NULL);
> > + break;
> > + }
> > + }
>
> Is the goal of this patch to bring up the display device early during
> boot?
Yes.
> If so, do we need to change pciinit.c? Could we instead add:
> if (!have_vga) try_setup_display_other();
> here and then let that function do all the work?
The only thing the pciinit patch effectively does (beside the printk) is
setting the PCI_COMMAND register. We could do that elsewhere. Either
the vgabios does it itself (are roms expected to do that?), or
try_setup_display_other() does it.
> Separately, I don't think we need to call vgahook_setup() if a real
> vga device wasn't found.
Indeed.
cheers,
Gerd
_______________________________________________
SeaBIOS mailing list
SeaBIOS@seabios.org
https://mail.coreboot.org/mailman/listinfo/seabios
