[Piglit] [PATCH] main: Test for CreateBuffers.

Laura Ekstrand laura at jlekstrand.net
Thu Mar 5 12:22:20 PST 2015


On Thu, Mar 5, 2015 at 12:49 AM, Martin Peres <martin.peres at linux.intel.com>
wrote:

> On 05/03/15 03:48, Ilia Mirkin wrote:
>
>> On Wed, Mar 4, 2015 at 8:38 PM, Laura Ekstrand <laura at jlekstrand.net>
>> wrote:
>>
>>> ---
>>>   tests/all.py                                       |  1 +
>>>   .../spec/arb_direct_state_access/CMakeLists.gl.txt |  1 +
>>>   .../spec/arb_direct_state_access/create-buffers.c  | 78
>>> ++++++++++++++++++++++
>>>   3 files changed, 80 insertions(+)
>>>   create mode 100644 tests/spec/arb_direct_state_access/create-buffers.c
>>>
>>> diff --git a/tests/all.py b/tests/all.py
>>> index 0f07105..e60a02c 100644
>>> --- a/tests/all.py
>>> +++ b/tests/all.py
>>> @@ -4006,6 +4006,7 @@ spec['ARB_direct_state_access']['compressedtexturesubimage']
>>> = PiglitGLTest(['ar
>>>   spec['ARB_direct_state_access']['texture-storage-multisample'] =
>>> PiglitGLTest(['arb_direct_state_access-texture-storage-multisample'],
>>> run_concurrent=True)
>>>   spec['ARB_direct_state_access']['texture-buffer'] =
>>> PiglitGLTest(['arb_direct_state_access-texture-buffer'],
>>> run_concurrent=True)
>>>   spec['ARB_direct_state_access']['texture-buffer-range'] =
>>> PiglitGLTest(['arb_direct_state_access-texture-buffer-range'],
>>> run_concurrent=True)
>>> +spec['ARB_direct_state_access']['create-buffers'] =
>>> PiglitGLTest(['arb_direct_state_access-create-buffers'],
>>> run_concurrent=True)
>>>   spec['ARB_direct_state_access']['namedbufferstorage-persistent'] =
>>> PiglitGLTest(['arb_direct_state_access-namedbufferstorage-persistent'],
>>> run_concurrent=True)
>>>   spec['ARB_direct_state_access']['namedbuffersubdata-vbo-sync'] =
>>> PiglitGLTest(['arb_direct_state_access-namedbuffersubdata-vbo-sync'],
>>> run_concurrent=True)
>>>   spec['ARB_direct_state_access']['clearnamedbufferdata-invalid-internal-format']
>>> = PiglitGLTest(['arb_direct_state_access-clearnamedbufferdata-invalid-internal-format'],
>>> run_concurrent=True)
>>> diff --git a/tests/spec/arb_direct_state_access/CMakeLists.gl.txt
>>> b/tests/spec/arb_direct_state_access/CMakeLists.gl.txt
>>> index b2b7465..2a4915d 100644
>>> --- a/tests/spec/arb_direct_state_access/CMakeLists.gl.txt
>>> +++ b/tests/spec/arb_direct_state_access/CMakeLists.gl.txt
>>> @@ -10,6 +10,7 @@ link_libraries (
>>>   )
>>>
>>>   piglit_add_executable (arb_direct_state_access-create-transformfeedbacks
>>> create-transformfeedbacks.c)
>>> +piglit_add_executable (arb_direct_state_access-create-buffers
>>> create-buffers.c)
>>>   piglit_add_executable (arb_direct_state_access-namedbufferstorage-persistent
>>> namedbufferstorage.c)
>>>   piglit_add_executable (arb_direct_state_access-namedbuffersubdata-vbo-sync
>>> namedbuffersubdata-vbo-sync.c)
>>>   piglit_add_executable (arb_direct_state_access-
>>> clearnamedbufferdata-invalid-internal-format
>>> clearnamedbufferdata-invalid-internal-format.c)
>>> diff --git a/tests/spec/arb_direct_state_access/create-buffers.c
>>> b/tests/spec/arb_direct_state_access/create-buffers.c
>>> new file mode 100644
>>> index 0000000..12fcf0b
>>> --- /dev/null
>>> +++ b/tests/spec/arb_direct_state_access/create-buffers.c
>>> @@ -0,0 +1,78 @@
>>> +/*
>>> + * Copyright 2014 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 create_buffers.c
>>> + *
>>> + * Tests glCreateBuffers to see if it behaves in the expected way,
>>> + * throwing the correct errors, etc.
>>> + */
>>> +
>>> +#include "piglit-util-gl.h"
>>> +
>>> +PIGLIT_GL_TEST_CONFIG_BEGIN
>>> +
>>> +       config.supports_gl_compat_version = 32;
>>>
>> This will never run on mesa, right? I assume you meant
>>
>> supports_gl_core_version = 31 (or 32?)
>>
>
> Why 32? 31 should be enough and that is also what I used on all my tests.

Oddly enough, there is no "core context" for 3.1.  It's not a thing.  31 is
strange like that.  Core contexts start at 3.2.  The reason why your tests
work with core_compat = 31 is because Piglit jumps through some hoops,
often promoting the user automatically to 32 or above behind the scenes.
If we are going to only test on core contexts, then we should probably
start at 32 since 31 is such a strange beast.

Now, as far as whether or not we should test on compat contexts, most of
the discussion is confusing to me.  Often DSA entry points work as long as
the system has the appropriate related extension (i.e. you can't call
CreateSamplers if you don't have samplers).  Whenever the battle over core
vs. compat and what version to use is finally decided, we will probably
have to go back and redo these statements in the tests. (Ouch :( )

>
>
>  +
>>> +       config.window_visual = PIGLIT_GL_VISUAL_RGBA |
>>> +               PIGLIT_GL_VISUAL_DOUBLE;
>>> +
>>> +PIGLIT_GL_TEST_CONFIG_END
>>> +
>>> +void
>>> +piglit_init(int argc, char **argv)
>>> +{
>>> +       piglit_require_extension("GL_ARB_direct_state_access");
>>> +}
>>> +
>>> +enum piglit_result
>>> +piglit_display(void)
>>> +{
>>> +       bool pass = true;
>>> +       GLuint name;
>>> +       GLboolean ret;
>>> +
>>> +       /* Throw some invalid inputs at glCreateBuffers. */
>>> +
>>> +       /* n is negative */
>>> +       glCreateBuffers(-1, &name);
>>> +       pass = piglit_check_gl_error(GL_INVALID_VALUE) && pass;
>>> +
>>> +       /* name is not a valid pointer */
>>> +       glCreateBuffers(1, 0);
>>> +       pass = piglit_check_gl_error(GL_NO_ERROR) && pass;
>>> +       glCreateBuffers(1, NULL);
>>> +       pass = piglit_check_gl_error(GL_NO_ERROR) && pass;
>>>
>>
> Good idea to tests this! No need to test for both 0 and NULL though.
>
Fixed.

>  +
>>> +
>>> +       /* Check if it actually generates a real buffer object */
>>> +
>>> +       glCreateBuffers(1, &name);
>>> +       pass = piglit_check_gl_error(GL_NO_ERROR) && pass;
>>> +
>>> +       ret = glIsBuffer(name);
>>> +       pass = (ret == GL_TRUE) && pass;
>>> +       pass = piglit_check_gl_error(GL_NO_ERROR) && pass;
>>> +
>>> +       return pass ? PIGLIT_PASS : PIGLIT_FAIL;
>>> +}
>>> +
>>> --
>>> 2.1.0
>>>
>>
> I would really like to have multiple subtests here to show what we really
> test and not to have to open the test c file to know what went wrong.
>
> Could you use the SUBTEST macros and give the same name to the tests that
> I had in all my create-* tests? [0]
>
Sure, that's done.

>
> Don't forget to delete the buffer at the end. Not that it should matter
> but some people apparently run valgrind / coverty on piglit and they find
> this kind of "bugs".''

Fixed:
http://cgit.freedesktop.org/~ldeks/piglit/commit/?h=adsa-framebuffers&id=826afe77a56d266680938f8eab6108c94e6e6622


>
> Thanks!
>
> [0] http://cgit.freedesktop.org/piglit/commit/?id=
> 0c8acebafefbe5cb400c055d9880788f22339ce3
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/piglit/attachments/20150305/40086277/attachment.html>


More information about the Piglit mailing list