RE: [virtio-dev] Re: [PATCH v2 00/20] Split device spec to its individual files

[Parav Pandit <parav@nvidia.com>]

> From: Cornelia Huck <cohuck@redhat.com>
> Sent: Tuesday, January 10, 2023 10:21 AM
> 
> On Tue, Jan 10 2023, Parav Pandit <parav@nvidia.com> wrote:
> 
> >> From: virtio-dev@lists.oasis-open.org
> >> <virtio-dev@lists.oasis-open.org> On Behalf Of Cornelia Huck
> >>
> >> On Mon, Jan 09 2023, "Michael S. Tsirkin" <mst@redhat.com> wrote:
> >> > Does makediff still work? Documentation says latexpand does not
> >> > support import. without latexdiff generating redlined versions
> >> > would be very difficult.
> >> >
> >> >
> >> > I am also worried about consistency since we already use \\input.
> >> > If using \\input means putting everything in a single directory,
> >> > that's a small price to pay:
> >> >
> >> > virtio-sound.tex + virtio-sound-conformance.tex
> >> >
> >> > is not fundamentally worse than
> >> > device-types/virtio-sound/device.tex
> >> > and device-types/virtio-sound/device-conformance.tex
> >> >
> >> > and it avoids the duplicated "device" in the name.
> >> >
> >> > Previously it looked like a cosmetic issue, but now it looks like
> >> > it's important.
> >>
> >> I agree, and we need to decide quickly what to do with the ballot. We
> >> don't want to merge v1, but the current votes still have a majority
> >> of 'yes'. My preference would be to withdraw the ballot, which needs
> >> to be done before
> >> 22:00 UTC today, if I'm not confused.
> >>
> >> Parav, what do you think? If you request to withdraw the ballot,
> >> that's easy to do; we'll just open a new one once we've agreed on a version.
> >
> > I am revising the v2 and should be available in 7 pm UTC time.
> > This will include,
> > a. white space removal at end of the net and blk files b. fix missing
> > device conformance links for 4 devices c. import to input d. continue
> > with directories e. rename device-types/<name>/device.tex to
> > device-types/<name>/description.tex
> 
> Apologies if I sound like a process stickler, but the main problem is that the
> ballot is currently about v1 of the patches (and we obviously can't change that
> while it is open, as that would invalidate the votes that already have been cast.)
If it auto invalidates, is there any withdrawal process needed?
If no, lets withdraw an re-vote on v3.

> If your v3 looks good, we need a new ballot to vote on that. As it stands now,
> the TC would have voted to include v1 with its known problems... that's why I
> think a withdrawal would be best.

We will use the same github issue for v3, and new ballot yes?