[Mesa-dev] [PATCH 03/20] t_dd_dmatmp: Disallow flat shading when rendering quad strips via tri strips

[Ville Syrjälä]

On Fri, May 15, 2015 at 12:08:33PM -0700, Ian Romanick wrote:
> On 03/23/2015 05:47 AM, ville.syrjala at linux.intel.com wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > 
> > When rendering quad strips via tri strips we can't get the provoking
> > vertex right, so disallow flat shading.
> 
> Same comments as for patch 2.

I'm not sure about piglit, but there's bunch of stuff in mesa-demos
that hit this stuff.

To clarify a bit, this series does fix the piglit provoking vertex
test on 855, but I'm unsure which patches precisely are the key to
that since a lot of them try to fix various provoking vertex issues.
I was too lazy to reverse bisect my own patches.

> 
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > ---
> >  src/mesa/tnl_dd/t_dd_dmatmp.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/src/mesa/tnl_dd/t_dd_dmatmp.h b/src/mesa/tnl_dd/t_dd_dmatmp.h
> > index 3ed4a98..f56b0aa 100644
> > --- a/src/mesa/tnl_dd/t_dd_dmatmp.h
> > +++ b/src/mesa/tnl_dd/t_dd_dmatmp.h
> > @@ -447,7 +447,7 @@ static void TAG(render_quad_strip_verts)( struct gl_context *ctx,
> >  
> >        FLUSH();
> >     }
> > -   else if (HAVE_TRI_STRIPS) {
> > +   else if (HAVE_TRI_STRIPS && ctx->Light.ShadeModel == GL_SMOOTH) {
> >        LOCAL_VARS;
> >        int dmasz = GET_SUBSEQUENT_VB_MAX_VERTS();
> >        int currentsz;
> > @@ -1124,7 +1124,7 @@ static GLboolean TAG(validate_render)( struct gl_context *ctx,
> >  	 } else if (HAVE_QUAD_STRIPS) {
> >  	    ok = GL_TRUE;
> >  	 } else {
> > -	    ok = HAVE_TRI_STRIPS;
> > +	    ok = (HAVE_TRI_STRIPS && ctx->Light.ShadeModel == GL_SMOOTH);
> >  	 }
> >  	 break;
> >        case GL_QUADS:
> > 

-- 
Ville Syrjälä
Intel OTC