[Piglit] [PATCH] Porting basic GL rendering paths test to piglit
Juliet Fru
julietfru at gmail.com
Fri Aug 28 05:54:43 PDT 2015
Hello IIia,
Thanks
On Fri, Aug 28, 2015 at 10:37 AM, Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> On Fri, Aug 28, 2015 at 4:48 AM, Juliet Fru <julietfru at gmail.com> wrote:
> > Hello Ilia,
> >
> >
> >
> >> > + }
> >> > +
> >> > + /* test always-pass paths */
> >> > + for (i = 0; i < num_paths; i++) {
> >> > + glClear(GL_COLOR_BUFFER_BIT);
> >> > +
> >> > + set_path_state(paths[i], ALWAYS_PASS);
> >> > +
> >> > + /* draw polygon */
> >> > + piglit_draw_rect(-1,-1,1,1);
> >> > +
> >> > + set_path_state(paths[i], DISABLE);
> >> > +
> >> > + /* test buffer */
> >> > + GLfloat pixel[3];
> >> > + glReadPixels(4, 4, 1, 1, GL_RGB, GL_FLOAT, pixel);
> >> > + if (pixel[0] != 1.0 || pixel[1] != 1.0 || pixel[2] !=
> >> > 1.0) {
> >> > + printf("\nFAIL: %s, should have had no
> effect"
> >> > + " (1, 1, 1) but actually modified the
> >> > fragment"
> >> > + " ( %f, %f, %f)\n",
> path_name(paths[i]),
> >> > pixel[0], pixel[1], pixel[2]);
> >> > + return PIGLIT_FAIL;
> >>
> >> You request a double-buffered visual but you never call
> >> piglit_present_results... afaik that'll end up with an unfilled window
> >> when you run without -fbo -auto... Please double-check that case.
> >>
> > I think I will have to replace the test buffer section with
> > piglit_probe_rect_rgb(), but you could please give a sample correction in
> > this case?
>
> Just add a call to piglit_present_results() before you return from the
> function.
>
> -ilia
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/piglit/attachments/20150828/321d085f/attachment.html>
More information about the Piglit
mailing list