What I don't understand is that all the lines starting with a "-" seems to be part of the GL_ARB_gpu_shader5 extension. See the line here: <a href="http://cgit.freedesktop.org/mesa/mesa/tree/docs/GL3.txt#n99">http://cgit.freedesktop.org/mesa/mesa/tree/docs/GL3.txt#n99</a> If I'm right, it means that, considering Ilia's web site, GL_ARB_gpu_shader5 is unsupported by R600, but everything in its sublist is supported. You see why it is confusing? <div class="gmail_quote">Le 29 sept. 2015 8:06 AM, "Marek Olšák" <<a href="mailto:maraeo@gmail.com">maraeo@gmail.com</a>> a écrit : <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">FMA isn't required really. R600 is mainly missing GS streams, which are complete on the mailing somewhere I think. Marek On Tue, Sep 29, 2015 at 7:32 AM, Romain Failliot <<a href="mailto:romain.failliot@foolstep.com">romain.failliot@foolstep.com</a>> wrote: > Hi! > > I'm diving into the unsupported extensions list and I'm wondering how is it > possible that GL_ARB_gpu_shader5 is unsupported for R600, but some of the > "sub-extensions" like "Dynamically uniform sampler array indices" are > supported nonetheless. > > That makes me wonder if "not done" sub-extensions, like "Fused > multiply-add", are really not done for R600 yet, or if they are indeed > unsupported as the parent extension status let us suppose. > > Thx > Romain > > 2015-07-31 23:42 GMT+02:00 Ilia Mirkin <<a href="mailto:imirkin@alum.mit.edu">imirkin@alum.mit.edu</a>>: >> >> OK, I believe I've fixed my list up. Note that you may have to >> shift-reload to get the updates, I think fd.o isn't setting the proper >> cache headers or something else is messed up. > > </blockquote></div>