On 06/05/2014 05:51 AM, Jose Fonseca wrote:
Looks ok. Though I still wonder if it wouldn't be to do as I suggested in another similar review request -- to treat config.window_width/height as "minimum" values, as opposed to maximum values. That is, something along the lines ofdiff --git a/tests/util/piglit-framework-gl.c b/tests/util/piglit-framework-gl.c index a97babc..aa91625 100644 --- a/tests/util/piglit-framework-gl.c +++ b/tests/util/piglit-framework-gl.c @@ -138,6 +138,15 @@ void piglit_gl_test_run(int argc, char *argv[], const struct piglit_gl_test_config *config) { +#ifdef _WIN32 + struct piglit_gl_test_config modified_config = *config; + if (modified_config.window_width < 160) + modified_config.window_width = 160; + if (modified_config.window_height < 160) + modified_config.window_height = 160; + config = &modified_config; +#endif
I think I'd remove the #ifdef _WIN32 check. If someone were to develop a test on Linux with a small window size and some hard-coded piglit_probe() positions or coordinate data (like these ext_transform_feedback tests) their test might work fine on Linux but then fail on Windows. I think it would be safer to do the same thing for all environments.
Anyway, when I get a little more time, I'll try out this approach. Thanks for reviewing! -Brian _______________________________________________ Piglit mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/piglit
