Re: [PATCH v20 2/2] vpci: translate virtual PCI bus topology for guests

[Jan Beulich <jbeulich@xxxxxxxx>]

On 18.04.2025 20:58, Stewart Hildebrand wrote:
> --- a/xen/drivers/vpci/vpci.c
> +++ b/xen/drivers/vpci/vpci.c
> @@ -174,6 +174,41 @@ int vpci_assign_device(struct pci_dev *pdev)
>  }
>  #endif /* __XEN__ */
>  
> +#ifdef CONFIG_HAS_VPCI_GUEST_SUPPORT
> +/*
> + * Find the physical device which is mapped to the virtual device
> + * and translate virtual SBDF to the physical one.
> + */
> +static const struct pci_dev *translate_virtual_device(const struct domain *d,
> +                                                      pci_sbdf_t *sbdf)
> +{
> +    const struct pci_dev *pdev;
> +
> +    ASSERT(!is_hardware_domain(d));
> +    ASSERT(rw_is_locked(&d->pci_lock));
> +
> +    for_each_pdev ( d, pdev )
> +    {
> +        if ( pdev->vpci && (pdev->vpci->guest_sbdf.sbdf == sbdf->sbdf) )
> +        {
> +            /* Replace guest SBDF with the physical one. */
> +            *sbdf = pdev->sbdf;
> +            return pdev;
> +        }
> +    }
> +
> +    return NULL;
> +}
> +#else
> +static const struct pci_dev *translate_virtual_device(const struct domain *d,
> +                                                      pci_sbdf_t *sbdf)
> +{
> +    ASSERT_UNREACHABLE();
> +
> +    return NULL;
> +}
> +#endif /* CONFIG_HAS_VPCI_GUEST_SUPPORT */

I'm not a maintainer of this code, but if I was I'd strictly request that the
#ifdef move inside the function, to limit redundancy. Even the "return NULL"
is common between both forms.

Jan