Re: [Qemu-devel] [PATCH RFC v6 04/20] virtio: add feature checking helpers

[Thomas Huth <thuth@linux.vnet.ibm.com>]

On Thu, 11 Dec 2014 14:25:06 +0100
Cornelia Huck <cornelia.huck@de.ibm.com> wrote:

> Add a helper function for checking whether a bit is set in the guest
> features for a vdev as well as one that works on a feature bit set.
> 
> Convert code that open-coded this: It cleans up the code and makes it
> easier to extend the guest feature bits.
> 
> Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
...
> diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
> index ef48550..56c92fb 100644
> --- a/hw/scsi/virtio-scsi.c
> +++ b/hw/scsi/virtio-scsi.c
> @@ -144,7 +144,7 @@ static int virtio_scsi_parse_req(VirtIOSCSIReq *req,
>       *
>       * TODO: always disable this workaround for virtio 1.0 devices.
>       */
> -    if ((vdev->guest_features & VIRTIO_F_ANY_LAYOUT) == 0) {
> +    if (!virtio_has_feature(vdev, VIRTIO_F_ANY_LAYOUT)) {

Wait ... this does not only look like a clean-up, but also like a
bug-fix to me, since it should have been "(1 << VIRTIO_F_ANY_LAYOUT)"
instead of "VIRTIO_F_ANY_LAYOUT" in the original code instead?

So in case this patch queue takes a little bit longer 'til it gets
upstream, do we might want to submit a separate patch for fixing this
issue first?

>          req_size = req->elem.out_sg[0].iov_len;
>          resp_size = req->elem.in_sg[0].iov_len;
>      }
> @@ -748,7 +748,7 @@ static void virtio_scsi_change(SCSIBus *bus, SCSIDevice *dev, SCSISense sense)
>      VirtIOSCSI *s = container_of(bus, VirtIOSCSI, bus);
>      VirtIODevice *vdev = VIRTIO_DEVICE(s);
> 
> -    if (((vdev->guest_features >> VIRTIO_SCSI_F_CHANGE) & 1) &&
> +    if (virtio_has_feature(vdev, VIRTIO_SCSI_F_CHANGE) &&
>          dev->type != TYPE_ROM) {
>          virtio_scsi_push_event(s, dev, VIRTIO_SCSI_T_PARAM_CHANGE,
>                                 sense.asc | (sense.ascq << 8));
> @@ -769,7 +769,7 @@ static void virtio_scsi_hotplug(HotplugHandler *hotplug_dev, DeviceState *dev,
>          blk_op_block_all(sd->conf.blk, s->blocker);
>      }
> 
> -    if ((vdev->guest_features >> VIRTIO_SCSI_F_HOTPLUG) & 1) {
> +    if (virtio_has_feature(vdev, VIRTIO_SCSI_F_HOTPLUG)) {
>          virtio_scsi_push_event(s, sd,
>                                 VIRTIO_SCSI_T_TRANSPORT_RESET,
>                                 VIRTIO_SCSI_EVT_RESET_RESCAN);
> @@ -783,7 +783,7 @@ static void virtio_scsi_hotunplug(HotplugHandler *hotplug_dev, DeviceState *dev,
>      VirtIOSCSI *s = VIRTIO_SCSI(vdev);
>      SCSIDevice *sd = SCSI_DEVICE(dev);
> 
> -    if ((vdev->guest_features >> VIRTIO_SCSI_F_HOTPLUG) & 1) {
> +    if (virtio_has_feature(vdev, VIRTIO_SCSI_F_HOTPLUG)) {
>          virtio_scsi_push_event(s, sd,
>                                 VIRTIO_SCSI_T_TRANSPORT_RESET,
>                                 VIRTIO_SCSI_EVT_RESET_REMOVED);
...
> diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
> index 2fede2e..f6c0379 100644
> --- a/include/hw/virtio/virtio.h
> +++ b/include/hw/virtio/virtio.h
> @@ -278,6 +278,17 @@ static inline void virtio_clear_feature(uint32_t *features, unsigned int fbit)
>      *features &= ~(1 << fbit);
>  }
> 
> +static inline bool __virtio_has_feature(uint32_t features, unsigned int fbit)
> +{
> +    assert(fbit < 32);
> +    return !!(features & (1 << fbit));
> +}
> +
> +static inline bool virtio_has_feature(VirtIODevice *vdev, unsigned int fbit)
> +{
> +    return __virtio_has_feature(vdev->guest_features, fbit);
> +}
> +

I've got to say that I'm a little bit unhappy with the naming of the
functions - and in contrast to the Linux kernel code, I think it is
also quite uncommon in the QEMU sources to use function names with
double underscores at the beginning.

Could you maybe rename the second function to "virtio_vdev_has_feature"
instead? And then remove the double underscores from the first function?

 Thomas