[Mesa-dev] [PATCH 0/2 v2] Add support for clip distances in Gallium

[Jose Fonseca]

----- Original Message -----
> On 12/13/2011 09:11 PM, Jose Fonseca wrote:
> > 
> > 
> > ----- Original Message -----
> >> This is an updated version of the patch set I sent to the list a
> >> few
> >> hours
> >> ago.
> > 
> > 
> >> There is now a TGSI property called
> >> TGSI_PROPERTY_NUM_CLIP_DISTANCES
> >> that drivers can use to determine how many of the 8 available clip
> >> distances
> >> are actually used by a shader.
> > 
> > Can't the info in TGSI_PROPERTY_NUM_CLIP_DISTANCES be easily
> > derived from the shader, and queried through
> > src/gallium/auxiliary/tgsi/tgsi_scan.h ?
> > 
> > 
> > Could you also elaborate on why TGSI_SEMANTIC_CLIPDIST is useful
> > for the drivers? I personally don't have nothing against it, but
> > just like to understand why it makes a difference.
> > 
> 
> Why does TGSI_SEMANTIC_*POSITION* make a difference ?
> 
> Right, because the position values are consumed by the fixed function
> rasterizer. So are the clip distances.
> 
> This is not about pipe_clip_state.ucp but about what this legacy
> cruft
> has to be turned into if GL_CLIP_PLANEi is used instead of GLSL 1.3's
> gl_ClipDistance[i].

I'm just surprised because I thought there was no more fixed function clipping.

Can you give an example of such legacy cruft?

> The same mentality ("What's information useful for ?") cost me
> TGSI_SEMANTIC_TEXCOORD and now I have to rely on a hack to make point
> coordinate replacement work on nvc0
> (http://cgit.freedesktop.org/mesa/mesa/tree/src/gallium/drivers/nvc0/nvc0_program.c#n29).
>
> I'm sorry I'm a bit sensitive on the issue of dropping information at
> the gallium threshold.

Please don't vent it on me.  As I explicitly said, I don't oppose (I imagined there was a good reason), and I'm asking for my own education.

And these are questions that always need to be asked.  Or even better, should be answered from the start -- commit messages in general, and interface changes in particular, should not just summarize what its being changed, but also _why_ it changed. The patch series included no pipe driver changes, so it's not possible even to induce from them.


Concerning TGSI_SEMANTIC_TEXCOORD I can't remember/comment if/what/why it was decided like that.  Please start a new thread and a link to any previous thread if you wanna revisit it.

I'm sorry the choices done in gallium cause you so much grief. But dropping information is a trade off between simple interface and a rich interface, and not inherently bad or good. The only question is have the right trade offs been made. And there's always room for improvement.


Jose