On Thu, Dec 17, 2015 at 11:40:39AM +0200, Marcel Apfelbaum wrote: > On 12/16/2015 09:48 PM, Eduardo Habkost wrote: > >On Tue, Dec 15, 2015 at 04:27:51PM +0200, Marcel Apfelbaum wrote: > >>On 12/11/2015 08:42 PM, Eduardo Habkost wrote: > >[...] > >>>@@ -131,8 +130,7 @@ static void pc_q35_init(MachineState *machine) > >>> rom_memory = get_system_memory(); > >>> } > >>> > >>>- guest_info = pc_guest_info_init(pcms); > >>>- guest_info->isapc_ram_fw = false; > >> > >>This may not be an issue, I just want be sure. > >>For Q35 isapc_ram_fw is always false, but now we are always querying > >>!pcmc->pci_enabled. > >> > >>Now we have a Q35 case when !pcmc->pci_enabled *can* be true. > > > >Do we? pcmc->pci_enabled is always true in Q35. > > OK, thanks, so all the pcmc->pci_enabled "if" clauses in > pc_q35_init are not necessary, right?
They are not necessary, but keeping them helps us identify duplicate code in pc_q35.c:pc_q35_init() and pc_piix.c:pc_init1() more easily (so we can move it to common code later). -- Eduardo
