qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 00/18] chardev: QOM cleanups


From: Daniel P . Berrangé
Subject: Re: [PATCH 00/18] chardev: QOM cleanups
Date: Fri, 11 Sep 2020 09:32:10 +0100
User-agent: Mutt/1.14.6 (2020-07-11)

On Fri, Sep 11, 2020 at 12:19:08PM +0400, Marc-André Lureau wrote:
> Hi
> 
> On Fri, Sep 11, 2020 at 12:10 PM Daniel P. Berrangé <berrange@redhat.com>
> wrote:
> 
> > On Fri, Sep 11, 2020 at 12:07:27PM +0400, Marc-André Lureau wrote:
> > > Hi
> > >
> > > On Thu, Sep 10, 2020 at 11:50 PM Eduardo Habkost <ehabkost@redhat.com>
> > > wrote:
> > >
> > > > Some chardev QOM cleanup patches had to be dropped from my queue
> > > > due to build erros introduced by code movements across ifdef
> > > > boundaries at char-parallel.c.  This series redo the changes from
> > > > those patches, but the macro renames are now a little different:
> > > >
> > > > In this version I have decided to rename the type checking macros
> > > > from *_CHARDEV to CHARDEV_* instead of renaming tye
> > > > TYPE_CHARDEV_* constants to TYPE_*_CHARDEV, to make the
> > > > identifiers actually match the QOM type name strings
> > > > ("chardev-*").
> > > >
> > >
> > > Sounds reasonable to me, but it loses the matching with the
> > > structure/object type name though.
> > >
> > > - MuxChardev *d = MUX_CHARDEV(s);
> > > + MuxChardev *d = CHARDEV_MUX(s);
> > >
> > > I have a preference for the first. Unless we rename all the chardev types
> > > too...
> >
> > I tend to think the structs should be renamed too - I've always considerd
> > them to be backwards.
> >
> > > Imho, the QOM type name is mostly an internal detail, the C type name is
> > > dominant in the code.
> >
> > Err it is the reverse. The QOM type name is exposed public API via QOM
> > commands, while the C struct names are a internal private detail.
> >
> 
> Yes, but without the chardev- prefix (unless you try object-add which I
> don't think will work with chardev)

Sure, that's just the way it had to be wired into the -chardev arg
syntax, but from the POV of exposed QOM type information the canonical
type name has the full chardev- prefix.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|




reply via email to

[Prev in Thread] Current Thread [Next in Thread]