Re: [Qemu-devel] [PATCH] virtio: avoid leading underscores for helpers

[Michael S. Tsirkin]

On Tue, Mar 10, 2015 at 05:03:47PM +0100, Cornelia Huck wrote:
> On Tue, 10 Mar 2015 15:27:24 +0100
> "Michael S. Tsirkin" <address@hidden> wrote:
> 
> > On Tue, Mar 10, 2015 at 03:22:29PM +0100, Cornelia Huck wrote:
> > > On Tue, 10 Mar 2015 15:12:14 +0100
> > > "Michael S. Tsirkin" <address@hidden> wrote:
> > > 
> > > > On Tue, Mar 10, 2015 at 02:32:15PM +0100, Cornelia Huck wrote:
> > > > > Commit ef546f1275f6563e8934dd5e338d29d9f9909ca6 ("virtio: add
> > > > > feature checking helpers") introduced a helper __virtio_has_feature.
> > > > > We don't want to use reserved identifiers, though, so let's
> > > > > rename __virtio_has_feature to virtio_has_feature and 
> > > > > virtio_has_feature
> > > > > to virtio_vdev_has_feature.
> > > > 
> > > > I don't think it's urgent to fix in master.
> > > > Let's focus on getting virtio 1.0 branch merged instead.
> > > 
> > > I stumbled over this actually when trying to update my virtio-1 branch.
> > > I already did that change there (as promised in
> > > <address@hidden>), but it got lost
> > > somewhere in my moving chaos.
> > > 
> > > What's the status of your virtio-1.0 branch?
> > 
> > virtio pci works there too now, so I started looking at upstreaming
> > stuff from that branch.  Already did some.
> > 
> > > Would it be worthwile for
> > > me to rebase on top of it so I can figure out which changes I have not
> > > yet sent out?
> > 
> > Absolutely.
> 
> OK, it's actually not that much:
> 
> - this change :)
> - All ccw accesses are BE (see
>   <address@hidden>). I'll do two
>   patches: One for the existing ccws which will go via my tree and one
>   for the new set-revision ccw which should be squashed into that patch.

Will rebasing virtio-1.0 on top of master after your patch
is upstream do the trick as well?

> - Use legacy/non-legacy feature bit getters instead of
>   revision-specific ones (see
>   <address@hidden>). Should probably
>   replace the existing patches introducing get_features_rev and using it
>   in virtio-blk.

Right, but for that, let's get it all in working order using patches on
top, first.  Then, re-split logically.

> Also, it seems there are some r-bs that had been given for my patches
> that are missing on your branch.

I might have missed some - can you hunt up the msg ids?