[Piglit] [PATCH] builtin_varyings: enable GL_POINT_SMOOTH

[Eric Anholt]

On Wed, 18 Jan 2012 22:29:39 +0400, Vadim Girlin <vadimgirlin at gmail.com> wrote:
> Otherwise point size clamping breaks gl_PointSize subtest
> ---
>  .../spec/ext_transform_feedback/builtin-varyings.c |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/tests/spec/ext_transform_feedback/builtin-varyings.c b/tests/spec/ext_transform_feedback/builtin-varyings.c
> index 1fec203..4aea344 100644
> --- a/tests/spec/ext_transform_feedback/builtin-varyings.c
> +++ b/tests/spec/ext_transform_feedback/builtin-varyings.c
> @@ -489,6 +489,7 @@ piglit_init(int argc, char **argv)
>  	glGenBuffers(1, &xfb_buf);
>  	glGenQueries(1, &query);
>  	glEnable(GL_VERTEX_PROGRAM_TWO_SIDE);
> +	glEnable(GL_POINT_SMOOTH);
>  }

Looking at the spec for point rasterization:

"If point size mode is enabled, then the derived point size is taken
From the (potentially clipped) shader built-in gl_PointSize written by
the geometry shader, or written by the vertex shader if no geometry
shader is active, and clamped to the implementation-dependent point size
range."

It looks like the clamping should happen in rasteriztaion, after
feedback.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/piglit/attachments/20120119/2df87e31/attachment.pgp>