[Mesa-dev] Mesa (d3d1x): d3d1x: add new Direct3D 10/11 COM state tracker for Gallium
José Fonseca
jfonseca at vmware.com
Tue Sep 21 05:28:45 PDT 2010
On Tue, 2010-09-21 at 05:00 -0700, Henri Verbeet wrote:
> On 21 September 2010 13:13, Luca Barbieri <luca at luca-barbieri.com> wrote:
> > Why are you claiming this?
> > I assume it's because of the comment in tpf.h, which states that it
> > has been written according to Microsoft's documentation, which is
> > available solely from reading the d3d11TokenizedProgramFormat.h header
> > in the DDK.
> > Using the header, which is documented in the DDK documentation as the
> > place to look for documentation of the format, as reference, doesn't
> > seem to me unusual or problematic.
> >
> ...
> > Could you please explain your concerns in more detail?
> >
> The basic rule Wine uses is that if you've ever seen Microsoft code,
> you can't work on similar code for Wine. That may or may not be more
> strict than would legally be required in a given jurisdiction, but
> Wine simply doesn't want to take that risk. I think Bridgman expressed
> the issue fairly well at some point as "Stay well back from the
> abyss".
>
> The only reason we can look at e.g. PSDK headers at all is because you
> can't copyright facts like constant values or structure layouts. Even
> then, we have to be very careful to only take the facts from the
> header and not anything that would be considered "expression".
> Non-trivial macros would probably be considered code. Wine certainly
> wouldn't go through all the trouble of figuring out the formats from
> scratch if we considered looking at the DDK implementation an
> acceptable risk. Also, at least a couple of years ago the license for
> the DDK explicitly restricted its use to Windows development. That's
> again something that may or may not hold up, but we're not going to
> try.
>
> Note that I've looked at neither the DDK nor your actual
> implementation. The fact that apparently you looked at the DDK while
> writing the code makes me cautious to avoid it.
Mesa never had such tight requirements: there are many contributors,
including full time professionals, which have seen a lot of stuff under
NDA, not to mention publicly available documents and reference code.
Nothing changed with Luca's contribution -- either you've been tainted
before, or you won't be tainted now.
But, even assuming for the sake of argument that Luca saw Microsoft
reference code, if his contribution is not derived code, then you
haven't seen any Microsoft code, and nothing prevents you to contribute
to whatever you want. WINE may refuse to integrate Luca's code directly,
but that's a different matter.
Jose
More information about the mesa-dev
mailing list