On 05/23/2012 06:03 PM, Brian Paul wrote: > On 05/23/2012 12:01 AM, Tapani Pälli wrote: >> introduce a new test which checks that GL_COLOR_MATERIAL state >> does not get lost with sequential queries. >> >> Signed-off-by: Tapani Pälli<[email protected]> >> --- >> tests/all.tests | 1 + >> tests/general/CMakeLists.gl.txt | 1 + >> tests/general/colormaterial-query.c | 80 >> +++++++++++++++++++++++++++++++++++ >> 3 files changed, 82 insertions(+), 0 deletions(-) >> create mode 100644 tests/general/colormaterial-query.c >> >> diff --git a/tests/all.tests b/tests/all.tests >> index 59d2983..5286f55 100644 >> --- a/tests/all.tests >> +++ b/tests/all.tests >> @@ -260,6 +260,7 @@ add_plain_test(general, 'dlist-fdo3129-01') >> add_plain_test(general, 'dlist-fdo3129-02') >> add_plain_test(general, 'dlist-fdo31590') >> add_plain_test(general, 'draw-arrays-colormaterial') >> +add_plain_test(general, 'colormaterial-query') >> add_plain_test(general, 'draw-batch') >> add_plain_test(general, 'draw-copypixels-sync') >> add_plain_test(general, 'draw-elements') >> diff --git a/tests/general/CMakeLists.gl.txt >> b/tests/general/CMakeLists.gl.txt >> index 2c93b56..5929d3b 100644 >> --- a/tests/general/CMakeLists.gl.txt >> +++ b/tests/general/CMakeLists.gl.txt >> @@ -31,6 +31,7 @@ piglit_add_executable (depth_clamp depth_clamp.c) >> piglit_add_executable (depthfunc depthfunc.c) >> piglit_add_executable (depthrange-clear depthrange-clear.c) >> piglit_add_executable (draw-arrays-colormaterial >> draw-arrays-colormaterial.c) >> +piglit_add_executable (colormaterial-query colormaterial-query.c) >> piglit_add_executable (draw-batch draw-batch.c) >> piglit_add_executable (draw-copypixels-sync draw-copypixels-sync.c) >> piglit_add_executable (draw-elements draw-elements.c) >> diff --git a/tests/general/colormaterial-query.c >> b/tests/general/colormaterial-query.c >> new file mode 100644 >> index 0000000..4934dca >> --- /dev/null >> +++ b/tests/general/colormaterial-query.c >> @@ -0,0 +1,80 @@ >> +/* >> + * Copyright © 2012 Intel Corporation >> + * >> + * Permission is hereby granted, free of charge, to any person >> obtaining a >> + * copy of this software and associated documentation files (the >> "Software"), >> + * to deal in the Software without restriction, including without >> limitation >> + * the rights to use, copy, modify, merge, publish, distribute, >> sublicense, >> + * and/or sell copies of the Software, and to permit persons to whom the >> + * Software is furnished to do so, subject to the following conditions: >> + * >> + * The above copyright notice and this permission notice (including >> the next >> + * paragraph) shall be included in all copies or substantial portions >> of the >> + * Software. >> + * >> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, >> EXPRESS OR >> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF >> MERCHANTABILITY, >> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT >> SHALL >> + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES >> OR OTHER >> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, >> ARISING >> + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR >> OTHER DEALINGS >> + * IN THE SOFTWARE. >> + * >> + */ >> + >> +/** >> + * \file colormaterial-query.c >> + * Test sequential calls to glMaterialfv and glGetMaterialfv >> + * >> + * This test exercises a Mesa bug where 2 sequential queries with >> + * glGetMaterialfv get wrong result. >> + * >> + * \author Tapani Pälli<[email protected]> >> + */ >> + >> +#include "piglit-util.h" >> + >> +int piglit_width = 100, piglit_height = 100; >> +int piglit_window_mode = GLUT_RGBA | GLUT_ALPHA | GLUT_DOUBLE; >> + >> +enum piglit_result >> +piglit_display(void) >> +{ >> + static const GLfloat ambient[4] = { 0.6f, 0.6f, 0.6f, 1.0f}; >> + static const GLfloat color[4] = { 1.0f, 0.0f, 0.0f, 1.0f}; >> + static GLfloat result[4]; >> + >> + GLboolean pass = GL_TRUE; >> + >> + glEnable(GL_COLOR_MATERIAL); > > You're not calling glColorMaterial() to indicate which material values > are tracking glColor. The defaults are GL_FRONT_AND_BACK / > GL_AMBIENT_AND_DIFFUSE. It might be nice to have a call to > glColorMaterial() even with the default values to make things a little > more obvious to the reader.
Yep, I originally had this but I did not want to do this as
glColorMaterial is not specified in OpenGL ES and I want the test to
work with that as well, also it is not necessary for triggering the bug.
>
>> + glEnable(GL_LIGHTING);
>
> I don't think you need to enable GL_LIGHTING for these material tests
> (unless that somehow factors into the original bug).
This is true, will remove, it was copy-paste remaining from the test I
created this one from :)
>
> The comments could be beefed up a bit...
>
>> + /* set a color */
>
> /* call glColor to set the front+back GL_AMBIENT materials */
>
>
>> + glColor4f(color[0], color[1], color[2], color[3]);
>> +
>> + /* 1st query that works */
>
> /* This glMaterial call should be a no-op since GL_AMBIENT is tracking
> glColor */
>
>> + glMaterialfv(GL_FRONT, GL_AMBIENT, ambient);
>> + glGetMaterialfv(GL_FRONT, GL_AMBIENT, (GLfloat*)result);
>> +
>> + if(memcmp(color, result, sizeof(&color)) != 0)
>
> Minor nit: we usually put a space between 'if' and '('.
>
>
>> + pass = GL_FALSE;
>> +
>> + /* 2nd query that fails if the bug exists */
>
> /* This glMaterial call should also be a no-op (see above) */
I'll fix these comments and the space issue.
>> + glMaterialfv(GL_FRONT, GL_AMBIENT, ambient);
>> + glGetMaterialfv(GL_FRONT, GL_AMBIENT, (GLfloat*)result);
>> +
>> + if(memcmp(color, result, sizeof(&color)) != 0)
>> + pass = GL_FALSE;
>> +
>> + glDisable(GL_LIGHTING);
>> + glDisable(GL_COLOR_MATERIAL);
>> +
>> + return pass ? PIGLIT_PASS : PIGLIT_FAIL;
>> +}
>> +
>> +
>> +void
>> +piglit_init(int argc, char **argv)
>> +{
>> + /* nop */
>> +}
>
> Ideally, a test called "colormaterial-query" would test querying all the
> different material attributes in various circumstances. But I guess
> this is OK. Others might feel differently.
Yep, it could iterate through different properties, it just feels a bit
unnecessary as this is really about specific bug and the bug would
happen with any property or with any face parameter.
> -Brian
Thanks for the review!
--
// Tapani
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Piglit mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/piglit
