Re: [PATCH v22 3/4] qcow2: add zstd cluster compression

[Eric Blake]

On 4/29/20 8:02 AM, Vladimir Sementsov-Ogievskiy wrote:

> > > > > +    /*
> > > > > +     * The compressed stream from the input buffer may consist of 
> > > > > more
> > > > > +     * than one zstd frame.
> > > >
> > > > Can it?
> > >
> > > If not, we must require it in the specification.
> >
> > Actually, now that you mention it, it would make sense anyway to add
> > some note to the specification on what exactly compressed with zstd 
> > means.

> So, we don't know do we want one frame restriction or not. Do you have a 
> preference?

I'm a fan of 'be strict in what you produce, liberal in what you 
accept'. While our implementation always produces only a single frame of 
compressed data, I think our decoder should be prepared to see more than 
one frame from other implementations, as that is more liberal than 
tightening the specification to require that encoding must produce 
exactly one frame.

--
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3226
Virtualization:  qemu.org | libvirt.org