bhyve: virtio: introduce vq_kick_enable() and vq_kick_disable()

[vmaffione]

Summary

The VirtIO standard supports two schemes for notification suppression: a notification enable bit and
a more sophisticated one (event_idx) that also supports delayed notifications.
Currently bhyve fully supports only the first scheme. This patch hides the notification
suppression internals by means of two inline routines, vq_kick_enable() and vq_kick_disable(),
and makes the code more readable.
Moreover, further improve readability by replacing the call to mb() with a call to
atomic_thread_fence_seq_cst(), which is already used in virtio.c

[pmooney_pfmooney.com]

usr.sbin/bhyve/pci_virtio_block.c

42 ↗

(On Diff #58461)

Worth adding the include to virtio.h rather than requiring it in each of the consumers?

[vmaffione]

usr.sbin/bhyve/pci_virtio_block.c

42 ↗

(On Diff #58461)

Agree, that was also my idea. But I observed that virtio.h does not include anything, so I assumed there was a reason for that.
Does anyone have objections on adding include <machine/atomic.h> in virtio.h?

[vmaffione]

Moved include <machine/atomic.h> to virtio.h.

[rgrimes]

In D20581#445096, @v.maffione_gmail.com wrote:

Moved include <machine/atomic.h> to virtio.h.

I am ok with this, but could you please produce your diff relative to /usr/src?

[vmaffione]

Provide patch relative to /usr/src/.