[Piglit] [PATCH 1/2] ext_transform_feedback: Fix default precision qualifiers

Timothy Arceri t_arceri at yahoo.com.au
Mon Aug 3 01:39:40 PDT 2015


On Mon, 2015-08-03 at 08:47 +0200, Iago Toral wrote:
> If no body has issues with these two patches I'd like to push them later
> this week.
> 
> Iago
> 
> On Tue, 2015-07-28 at 13:52 +0200, Iago Toral Quiroga wrote:
> > Default ES precision qualifiers for integer types in the VS and FS
> > are different and that should make the test fail to compile/link
> > if Mesa actually checked for these things (because it would realize
> > that VS outputs and FS inputs do not have the exact same type as
> > a consequence).

I don't think this is correct. The GLES Shader Language spec says:

"Precision qualifiers for outputs in one shader matched to inputs in another
shader need not match when
both shaders are linked into the same program. When both shaders are in
separate programs, mismatched 
precision qualifiers will result in a program interface mismatch that will
result in program pipeline 
validation failures, as described in section 7.4.1 (“Shader Interface
Matching”) of the OpenGL ES 3.1 
Specification."

So it seems they only need to match for SSO (which these piglit tests don't
make use of).
> > 
> > Fix this by setting a default precision qualifier for integers.
> > ---
> >  tests/spec/ext_transform_feedback/structs.c | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/tests/spec/ext_transform_feedback/structs.c 
> > b/tests/spec/ext_transform_feedback/structs.c
> > index 3d90bb9..c754b9f 100644
> > --- a/tests/spec/ext_transform_feedback/structs.c
> > +++ b/tests/spec/ext_transform_feedback/structs.c
> > @@ -126,6 +126,7 @@ PIGLIT_GL_TEST_CONFIG_END
> >  static const char gles3_header[] =
> >  	"#version 300 es\n"
> >  	"precision highp float;\n"
> > +	"precision highp int;\n"
> >  	"#define DECLARE_VARYING(DIR, TYPE, NAME) flat DIR TYPE NAME\n"
> >  	"#define VARYING(NAME) NAME\n";
> >  
> 
> 
> _______________________________________________
> Piglit mailing list
> Piglit at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/piglit


More information about the Piglit mailing list