On Wed, Jan 24, 2024 at 20:37:39 +0100, Andrea Bolognani wrote:
> Extract the logic from qemuDomainControllerDefPostParse().
> 
> The code is mostly unchanged, but there's a subtle difference:
> the piix3-uhci has been moved from the top of the chunk to the
> bottom. This is because the original code set cont->model
> directly, which made it okay to start with a suboptimal default
> and subsequently overwrite it with a better one; now that we
> return the selected value instead, we need to make sure that
> we only ever consider piix3-uhci if everything else has failed.
> 
> Signed-off-by: Andrea Bolognani <abolo...@redhat.com>
> ---
>  src/qemu/qemu_domain.c | 100 +++++++++++++++++++++++------------------
>  1 file changed, 56 insertions(+), 44 deletions(-)
> 
> diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
> index 7475fb4f39..09f572b0b5 100644
> --- a/src/qemu/qemu_domain.c
> +++ b/src/qemu/qemu_domain.c
> @@ -4151,6 +4151,61 @@ qemuDomainDefaultSCSIControllerModel(const 
> virDomainDef *def,
>  }
>  
>  
> +static int
> +qemuDomainDefaultUSBControllerModel(const virDomainDef *def,
> +                                    const virDomainControllerDef *cont,
> +                                    virQEMUCaps *qemuCaps,
> +                                    unsigned int parseFlags)
> +{
> +    /* Pick a suitable default model for the USB controller if none
> +     * has been selected by the user and we have the qemuCaps for
> +     * figuring out which controllers are supported.
> +     *
> +     * We rely on device availability instead of setting the model
> +     * unconditionally because, for some machine types, there's a
> +     * chance we will get away with using the legacy USB controller
> +     * when the relevant device is not available.
> +     *
> +     * See qemuBuildControllersCommandLine() */
> +
> +    if (ARCH_IS_S390(def->os.arch)) {
> +        if (cont->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE) {
> +            /* set the default USB model to none for s390 unless an
> +             * address is found */
> +            return VIR_DOMAIN_CONTROLLER_MODEL_USB_NONE;
> +        }
> +    } else if (ARCH_IS_PPC64(def->os.arch)) {
> +        /* To not break migration we need to set default USB controller
> +         * for ppc64 to pci-ohci if we cannot change ABI of the VM.
> +         * The nec-usb-xhci or qemu-xhci controller is used as default
> +         * only for newly defined domains or devices. */
> +        if ((parseFlags & VIR_DOMAIN_DEF_PARSE_ABI_UPDATE) &&
> +            virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE_QEMU_XHCI)) {
> +            return VIR_DOMAIN_CONTROLLER_MODEL_USB_QEMU_XHCI;
> +        } else if ((parseFlags & VIR_DOMAIN_DEF_PARSE_ABI_UPDATE) &&
> +            virQEMUCapsGet(qemuCaps, QEMU_CAPS_NEC_USB_XHCI)) {
> +            return VIR_DOMAIN_CONTROLLER_MODEL_USB_NEC_XHCI;
> +        } else if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_PCI_OHCI)) {
> +            return VIR_DOMAIN_CONTROLLER_MODEL_USB_PCI_OHCI;
> +        } else {
> +            /* Explicitly fallback to legacy USB controller for PPC64. */
> +            return -1;

So this is extracting the logic of passing
VIR_DOMAIN_CONTROLLER_MODEL_USB_DEFAULT as model, which is -1, thus at
this point faithful representation what the code did. The very next
commit though declares -1 to be an error in the function comment which
is not true.

Extract it here as the proper enum value.

> +        }
> +    } else if (def->os.arch == VIR_ARCH_AARCH64) {
> +        if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE_QEMU_XHCI))
> +            return VIR_DOMAIN_CONTROLLER_MODEL_USB_QEMU_XHCI;
> +        else if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_NEC_USB_XHCI))
> +            return VIR_DOMAIN_CONTROLLER_MODEL_USB_NEC_XHCI;
> +    }
> +
> +    /* Default USB controller is piix3-uhci if available. */
> +    if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_PIIX3_USB_UHCI))
> +        return VIR_DOMAIN_CONTROLLER_MODEL_USB_PIIX3_UHCI;
> +
> +    return -1;

This is later on documented via a comment but also should use the proper
enum value.

Reviewed-by: Peter Krempa <pkre...@redhat.com>
_______________________________________________
Devel mailing list -- devel@lists.libvirt.org
To unsubscribe send an email to devel-le...@lists.libvirt.org

Reply via email to