Hi Cornelia,
Thanks for the insightful comments and ideas. Answers below:
On lun, ago 08 2022 at 15:47:06, Cornelia Huck <cohuck@xxxxxxxxxx> wrote:
> Hm... so I'm not quite happy with those two paragraphs, but I'm not sure
> how to make it better.
>
> - While the origins of virtio are hypervisor implementations, the
> standard is describing the mechanisms of device <-> driver
> communication, regardless whether the device is a hardware entity or
> something emulated by a hypervisor.
> - I'm not quite sure what you are referring to with "real host
> devices". We can have e.g. a real disk that is handed to a guest via
> virtio-blk as a whole, or we can have some kind of file that is
> exposed via virtio-blk. Other device types can also be completely
> emulated.
> - The OS picks the driver depending on what device is discovers; virtio
> device drivers are not any different from other device drivers in that
> regard.
>
> So I think the key pieces of virtio are the following:
>
> - it is an open standard
> - it describes a protocol, which can be used to implement a lot of
> different device types
> - those devices are exposed to the operating system via standard
> mechanisms such as PCI
> - virtio devices can be implemented in various ways, such as in
> hypervisors (more common) or as a real hardware device
>
> For the remainder of this document, it is probably fine to focus on the
> hypervisor use case.
How about this as an introduction?
Virtio is an open standard that defines a protocol to communicate
drivers and devices of different types, see Chapter 5 ("Device
Types") of the virtio spec `[1]`_. Originally developed as a
standard for paravirtualized devices implemented by a hypervisor, it
can be used to interface any compliant device (real or emulated)
with a driver.
For illustrative purposes, this document will focus on the common
case of a Linux kernel running in a virtual machine and using
paravirtualized devices provided by the hypervisor, which exposes
them as virtio devices via standard mechanisms such as PCI.
> s/in/on/ ?
> s/RedHat/Red Hat/ :)
Got it.
> See my comments above, virtio devices can also be fully emulated or real
> hardware devices.
>
> [...]
>
> I think we should not make any statement regarding frequency of new
> additions; sometimes, there's a flurry of activity, sometimes, there's
> nothing for ages :)
Ok, I'll simplify the intro to this:
This document serves as a basic guideline for driver programmers
that need to hack a new virtio driver or understand the essentials
of the existing ones. See :ref:`Virtio on Linux <virtio>` for a
general overview of virtio.
> s/should/needs to/ ?
Ack.
> s/defined/added/ ?
Ack.
> Maybe "Device ids need to be at least reserved in the virtio spec before
> being added to that file." ?
Sure, I'll add all of these to the next version.
Cheers,
Ricardo
