On 03/04/19 20:48, Philippe Mathieu-Daudé wrote:
> PCMachineState will be used in the next patch.
> Since We can access PCMachineClass from it, directly use it.
> We can also access pcmc->pci_enabled, remove the isapc_ram_fw
> argument.
>
> Signed-off-by: Markus Armbruster <arm...@redhat.com>
> [PMD: Extracted from bigger patch]
> Signed-off-by: Philippe Mathieu-Daudé <phi...@redhat.com>
> ---
> hw/i386/pc.c | 2 +-
> hw/i386/pc_sysfw.c | 8 +++++---
> include/hw/i386/pc.h | 3 +--
> 3 files changed, 7 insertions(+), 6 deletions(-)
>
> diff --git a/hw/i386/pc.c b/hw/i386/pc.c
> index 3889eccdc3..3cd8ed1794 100644
> --- a/hw/i386/pc.c
> +++ b/hw/i386/pc.c
> @@ -1830,7 +1830,7 @@ void pc_memory_init(PCMachineState *pcms,
> }
>
> /* Initialize PC system firmware */
> - pc_system_firmware_init(rom_memory, !pcmc->pci_enabled);
> + pc_system_firmware_init(pcms, rom_memory);
>
> option_rom_mr = g_malloc(sizeof(*option_rom_mr));
> memory_region_init_ram(option_rom_mr, NULL, "pc.rom", PC_ROM_SIZE,
> diff --git a/hw/i386/pc_sysfw.c b/hw/i386/pc_sysfw.c
> index 46b87afe23..55a3c563df 100644
> --- a/hw/i386/pc_sysfw.c
> +++ b/hw/i386/pc_sysfw.c
> @@ -231,15 +231,17 @@ static void old_pc_system_rom_init(MemoryRegion
> *rom_memory, bool isapc_ram_fw)
> bios);
> }
>
> -void pc_system_firmware_init(MemoryRegion *rom_memory, bool isapc_ram_fw)
> +void pc_system_firmware_init(PCMachineState *pcms,
> + MemoryRegion *rom_memory)
> {
> + PCMachineClass *pcmc = PC_MACHINE_GET_CLASS(pcms);
> DriveInfo *pflash_drv;
>
> pflash_drv = drive_get(IF_PFLASH, 0, 0);
>
> - if (isapc_ram_fw || pflash_drv == NULL) {
> + if (!pcmc->pci_enabled || pflash_drv == NULL) {
> /* When a pflash drive is not found, use rom-mode */
> - old_pc_system_rom_init(rom_memory, isapc_ram_fw);
> + old_pc_system_rom_init(rom_memory, true);
> return;
> }
>
This code extraction (as a pure refactoring) is incorrect.
In Markus's patch at
<https://lists.gnu.org/archive/html/qemu-devel/2019-02/msg06804.html>,
the condition that guards the old_pc_system_rom_init() call loses one
half of the disjunction:
- if (isapc_ram_fw || pflash_drv == NULL) {
+ if (!pcmc->pci_enabled) {
namely the "pflash_drv == NULL" sub-condition. Therefore the
old_pc_system_rom_init() call can hardwire the second argument as "true":
- old_pc_system_rom_init(rom_memory, isapc_ram_fw);
+ old_pc_system_rom_init(rom_memory, true);
However, in this refactoring, the (pflash_drv == NULL) subcondition is
not deleted; only the original condition is expressed differently.
Therefore we can't replace the "isapc_ram_fw" argument of
old_pc_system_rom_init() with plain "true", we have to use
"!pcmc->pci_enabled" instead.
However, if we write down "!pcmc->pci_enabled" twice for this purpose,
then it becomes too complex to read (for me anyway). So in that case I'd
even suggest keeping the "isapc_ram_fw" local variable, just turn it
into a normal local variable rather than take it as a parameter.
In the next patch, "isapc_ram_fw" can be killed just the same.
Thanks
Laszlo
> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> index 3ff127ebd0..eb7360feac 100644
> --- a/include/hw/i386/pc.h
> +++ b/include/hw/i386/pc.h
> @@ -278,8 +278,7 @@ extern PCIDevice *piix4_dev;
> int piix4_init(PCIBus *bus, ISABus **isa_bus, int devfn);
>
> /* pc_sysfw.c */
> -void pc_system_firmware_init(MemoryRegion *rom_memory,
> - bool isapc_ram_fw);
> +void pc_system_firmware_init(PCMachineState *pcms, MemoryRegion *rom_memory);
>
> /* acpi-build.c */
> void pc_madt_cpu_entry(AcpiDeviceIf *adev, int uid,
>
