On Fri, 14 Oct 2022 at 12:10, Philippe Mathieu-Daudé
<phi...@fungible.com> wrote:
>
> From: David Daney <david.da...@fungible.com>
>
> Most other virtio-pci devices allow MSI-X, let's have it for rng too.
>
> Signed-off-by: David Daney <david.da...@fungible.com>
> Reviewed-by: Marcin Nowakowski <marcin.nowakow...@fungible.com>
> Signed-off-by: Philippe Mathieu-Daudé <phi...@fungible.com>
> ---
> hw/virtio/virtio-rng-pci.c | 14 ++++++++++++++
> 1 file changed, 14 insertions(+)
>
> diff --git a/hw/virtio/virtio-rng-pci.c b/hw/virtio/virtio-rng-pci.c
> index 151ece6f94..6e76f8b57b 100644
> --- a/hw/virtio/virtio-rng-pci.c
> +++ b/hw/virtio/virtio-rng-pci.c
> @@ -13,6 +13,7 @@
>
> #include "hw/virtio/virtio-pci.h"
> #include "hw/virtio/virtio-rng.h"
> +#include "hw/qdev-properties.h"
> #include "qapi/error.h"
> #include "qemu/module.h"
> #include "qom/object.h"
> @@ -31,11 +32,23 @@ struct VirtIORngPCI {
> VirtIORNG vdev;
> };
>
> +static Property virtio_rng_properties[] = {
> + DEFINE_PROP_BIT("ioeventfd", VirtIOPCIProxy, flags,
> + VIRTIO_PCI_FLAG_USE_IOEVENTFD_BIT, true),
ioeventfd is unrelated to MSI-X. It controls whether the virtqueue
kick is handled in the vCPU thread (ioctl(KVM_RUN) exit) or the main
loop thread (KVM ioeventfd).
I took a quick look at virtio-rng.c to see if there is anything that
doesn't work in the main loop. It looks okay to me but consider
dropping this line to minimize the risk of breaking something.
> + DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors,
> + DEV_NVECTORS_UNSPECIFIED),
> + DEFINE_PROP_END_OF_LIST(),
> +};
> +
> static void virtio_rng_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp)
> {
> VirtIORngPCI *vrng = VIRTIO_RNG_PCI(vpci_dev);
> DeviceState *vdev = DEVICE(&vrng->vdev);
>
> + if (vpci_dev->nvectors == DEV_NVECTORS_UNSPECIFIED) {
> + vpci_dev->nvectors = 2;
Checked against the spec: 1 configuration change notification and 1
virtqueue used buffer notification = 2 MSI-X vectors.
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
> + }
> +
> if (!qdev_realize(vdev, BUS(&vpci_dev->bus), errp)) {
> return;
> }
> @@ -54,6 +67,7 @@ static void virtio_rng_pci_class_init(ObjectClass *klass,
> void *data)
> pcidev_k->device_id = PCI_DEVICE_ID_VIRTIO_RNG;
> pcidev_k->revision = VIRTIO_PCI_ABI_VERSION;
> pcidev_k->class_id = PCI_CLASS_OTHERS;
> + device_class_set_props(dc, virtio_rng_properties);
> }
>
> static void virtio_rng_initfn(Object *obj)
> --
> 2.37.3
>
>
