Il 25/07/2012 00:33, Nicholas A. Bellinger ha scritto:
> +int event_notifier_notify(EventNotifier *e)
> +{
> + uint64_t value = 1;
> + int r;
> +
> + assert(event_notifier_valid(e));
> + r = write(e->fd, &value, sizeof(value));
> + if (r < 0) {
> + return -errno;
> + }
> + assert(r == sizeof(value));
> + return 0;
> +}
Note we now have event_notifier_set.
> +#define EVENT_NOTIFIER_INITIALIZER ((EventNotifier){ .fd = -1 })
This is problematic when your event notifier is inside a struct, because
it is extremely easy to forget the initializer. You have to initialize
them yourself. Also, the right thing to test is not whether the
notifier is initialized; it is whether the notifier is actually checked
in QEMU's select() loop.
So, I would prefer avoiding event_notifier_valid and just use a boolean
(in virtio_queue_set_host_notifier_fd_handler and
virtio_queue_set_guest_notifier_fd_handler) to track whether the
notifiers are in use.
Paolo
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
