[Mesa-dev] talloc (Was: Merge criteria for glsl2 branch)

[José Fonseca]

On Thu, 2010-08-12 at 08:33 -0700, Matt Turner wrote:
> On Thu, Aug 12, 2010 at 11:27 AM, José Fonseca <jfonseca at vmware.com> wrote:
> > On Thu, 2010-08-12 at 07:58 -0700, Matt Turner wrote:
> >> On Thu, Aug 12, 2010 at 7:00 AM, José Fonseca <jfonseca at vmware.com> wrote:
> >> >> Really optimising for the wrong bunch of people here by dragging this
> >> >> stuff into mesa git.
> >> >
> >> > Many projects do this: they include the source of other projects, to
> >> > make it easier to build without having to build all dependencies.
> >>
> >> This is true, but it's also quite annoying. Take ioquake3 for example,
> >> in order to make it simpler for Windows and Mac OS developers, they
> >> include in SVN things like the libSDL headers, OpenAL headers, static
> >> libraries for OpenAL, libSDL, and libcurl, and the sources for
> >> libspeex, libcurl, and zlib.
> >>
> >> So the Linux packagers have to jump through hoops trying to untangle this mess.
> >>
> >> What Dave's saying with "optimizing for the wrong people" is that
> >> we're including lots of things in the Mesa code that most people (most
> >> users of Mesa are Linux/BSD/UNIX users) don't need, and by including
> >> these things with Mesa, we're making it more difficult to package Mesa
> >> for most people.
> >>
[...]
> 
> That's not really my intention. I think we can easily end the whole
> discussion just by moving the build dependencies that are used for
> Windows to a separate repository. It shouldn't make things more
> difficult for the VMware guys (I wouldn't think, at least) and it
> would make things cleaner for the Linux guys.

OK.

What about this:

For GLUT, GLEW, LLVM and all other dependencies I'll just make a SDK
with the binaries, with debug & release, 32 & 64 bit, MinGW & MSVC
versions. One seldom needs to modify the source anyway, and they have
active upstream development.

But I perceive talloc as different from all above: it's very low level
and low weight library, providing very basic functionality, and upstream
never showed interest for Windows portability. I'd really prefer to see
the talloc source bundled (and only compiled on windows), as a quick way
to have glsl2 merged without causing windows build failures. 

(Hopefully in the future we could have jakob Bornecrantz or Aras
Pranckevičius re-implement a BSD-like version of it, and therefore
eliminating all the different licensing concerns.)

Jose