Re: [libvirt PATCH 2/2] qemu: support use of stateless EFI firmware

[Andrea Bolognani <abologna@xxxxxxxxxx>]

On Fri, Jul 22, 2022 at 05:23:17PM +0100, Daniel P. Berrangé wrote:
>          VIR_DEBUG("decided on firmware '%s' template '%s' NVRAM '%s'",
>                    def->os.loader->path,
> -                  def->os.loader->nvramTemplate,
> -                  def->os.loader->nvram->path);
> +                  NULLSTR(def->os.loader->nvramTemplate),
> +                  NULLSTR(def->os.loader->nvram ? def->os.loader->nvram->path : NULL));

It would be nice to reflect in the live XML whether or not a
stateless firmware has been picked. Something along the lines of the
hastily thrown together, very lightly tested diff below should do the
trick.


diff --git a/src/qemu/qemu_firmware.c b/src/qemu/qemu_firmware.c
index eb7abb0b32..68d562285e 100644
--- a/src/qemu/qemu_firmware.c
+++ b/src/qemu/qemu_firmware.c
@@ -1200,6 +1200,9 @@ qemuFirmwareEnableFeatures(virQEMUDriver *driver,
                 def->os.loader->nvram->format = VIR_STORAGE_FILE_RAW;
                 qemuDomainNVRAMPathFormat(cfg, def,
&def->os.loader->nvram->path);
             }
+            def->os.loader->stateless = VIR_TRISTATE_BOOL_NO;
+        } else if (flash->mode == QEMU_FIRMWARE_FLASH_MODE_STATELESS) {
+            def->os.loader->stateless = VIR_TRISTATE_BOOL_YES;
         }

         VIR_DEBUG("decided on firmware '%s' template '%s' NVRAM '%s'",
-- 
Andrea Bolognani / Red Hat / Virtualization