[Piglit] [PATCH] ext_framebuffer_blit: Verify framebuffer bindings after calling glGenerateMipmap
From: Ian RomanickMesa's meta path for glGenerateMipmap had a problem that it assumed the two bindings were the same. As a result, one of them was incorrect after returning from _mesa_meta_GenerateMipmap. NOTE: This currently fails on all Mesa drivers that use _mesa_meta_GenerateMipmap. Signed-off-by: Ian Romanick --- tests/all.py | 3 +- .../spec/arb_framebuffer_object/CMakeLists.gl.txt | 1 + .../GenerateMipmap-versus-READ_FRAMEBUFFER.c | 129 + 3 files changed, 132 insertions(+), 1 deletion(-) create mode 100644 tests/spec/arb_framebuffer_object/GenerateMipmap-versus-READ_FRAMEBUFFER.c diff --git a/tests/all.py b/tests/all.py index f520920..431ea30 100644 --- a/tests/all.py +++ b/tests/all.py @@ -2003,7 +2003,7 @@ with profile.group_manager( g(['fbo-generatemipmap-1d']) g(['fbo-generatemipmap-1d', 'RGB9_E5']) g(['fbo-generatemipmap-3d']) -g(['fbo-generatemipmap-3d', 'RGB9_E5']) + for format in ('rgba', 'depth', 'stencil'): for test_mode in ('draw', 'read'): g(['framebuffer-blit-levels', test_mode, format], @@ -2705,6 +2705,7 @@ with profile.group_manager( g(['fbo-srgb-blit']) g(['fbo-sys-blit'], run_concurrent=False) g(['fbo-sys-sub-blit'], run_concurrent=False) +g(['fbo-generatemipmap-versus-READ_FRAMEBUFFER']) with profile.group_manager( PiglitGLTest, diff --git a/tests/spec/arb_framebuffer_object/CMakeLists.gl.txt b/tests/spec/arb_framebuffer_object/CMakeLists.gl.txt index 06d5978..e475e9e 100644 --- a/tests/spec/arb_framebuffer_object/CMakeLists.gl.txt +++ b/tests/spec/arb_framebuffer_object/CMakeLists.gl.txt @@ -18,5 +18,6 @@ piglit_add_executable(same-attachment-glFramebufferRenderbuffer-GL_DEPTH_STENCIL piglit_add_executable (fbo-blit-scaled-linear blit-scaled-linear.cpp) piglit_add_executable (fbo-attachments-blit-scaled-linear attachments-blit-scaled-linear.cpp) piglit_add_executable(fbo-drawbuffers-none fbo-drawbuffers-none.c) +piglit_add_executable(fbo-generatemipmap-versus-READ_FRAMEBUFFER GenerateMipmap-versus-READ_FRAMEBUFFER.c) # vim: ft=cmake: diff --git a/tests/spec/arb_framebuffer_object/GenerateMipmap-versus-READ_FRAMEBUFFER.c b/tests/spec/arb_framebuffer_object/GenerateMipmap-versus-READ_FRAMEBUFFER.c new file mode 100644 index 000..c46f323 --- /dev/null +++ b/tests/spec/arb_framebuffer_object/GenerateMipmap-versus-READ_FRAMEBUFFER.c @@ -0,0 +1,129 @@ +/* + * Copyright © 2015 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 GenerateMipmap-versus-READ_FRAMEBUFFER.c + * Verify that GL_READ_FRAMEBUFFER_BINDING and GL_DRAW_FRAMEBUFFER_BINDING are + * correct after calling glGenerateMipmap. + * + * Mesa's meta path for glGenerateMipmap had a problem that it assumed the two + * bindings were the same. As a result, one of them was incorrect after + * returning from _mesa_meta_GenerateMipmap. + */ + +#include "piglit-util-gl.h" + +PIGLIT_GL_TEST_CONFIG_BEGIN + + config.supports_gl_compat_version = 10; + config.window_visual = PIGLIT_GL_VISUAL_RGB; + +PIGLIT_GL_TEST_CONFIG_END + +static const GLuint texels[16] = { + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0 +}; + +void +piglit_init(int argc, char **argv) +{ + GLuint fbos[2]; + GLuint texture; + GLuint draw_binding; + GLuint read_binding; + bool pass = true; + + if (piglit_get_gl_version() >= 30 || + piglit_is_extension_supported("GL_ARB_framebuffer_object")) { + glGenFramebuffers(2, fbos); + glBindFramebuffer(GL_DRAW_FRAMEBUFFER, fbos[0]); + glBindFramebuffer(GL_READ_FRAMEBUFFER, fbos[1]); + } else if (piglit_is_extension_supported("GL_EXT_framebuffer_object") && +
Re: [Piglit] [PATCH 1/6] shader_runner: Check feature support before querying GL_MAX_*.
On Wed, Nov 11, 2015 at 4:47 PM, Kenneth Graunkewrote: > On Tuesday, November 10, 2015 10:46:18 PM Matt Turner wrote: >> Otherwise, these will generate an error (to be noticed sometime later >> when glGetError() is called), often resulting in a test failure. >> --- >> tests/shaders/shader_runner.c | 18 -- >> 1 file changed, 12 insertions(+), 6 deletions(-) >> >> diff --git a/tests/shaders/shader_runner.c b/tests/shaders/shader_runner.c >> index 32ac7bd..4597b46 100644 >> --- a/tests/shaders/shader_runner.c >> +++ b/tests/shaders/shader_runner.c >> @@ -3111,12 +3111,18 @@ piglit_init(int argc, char **argv) >> if (piglit_get_gl_version() >= 32) >> glGetIntegerv(GL_MAX_VERTEX_OUTPUT_COMPONENTS, >> _max_vertex_output_components); >> - glGetIntegerv(GL_MAX_FRAGMENT_UNIFORM_COMPONENTS, >> - _max_fragment_uniform_components); >> - glGetIntegerv(GL_MAX_VERTEX_UNIFORM_COMPONENTS, >> - _max_vertex_uniform_components); >> - glGetIntegerv(GL_MAX_VARYING_COMPONENTS, >> - _max_varying_components); >> + if (piglit_get_gl_version() >= 20 || >> + piglit_is_extension_supported("GL_ARB_fragment_shader")) >> + glGetIntegerv(GL_MAX_FRAGMENT_UNIFORM_COMPONENTS, >> + _max_fragment_uniform_components); >> + if (piglit_get_gl_version() >= 20 || >> + piglit_is_extension_supported("GL_ARB_vertex_shader")) >> + glGetIntegerv(GL_MAX_VERTEX_UNIFORM_COMPONENTS, >> + _max_vertex_uniform_components); > > Above this, we call piglit_require_GLSL()...it seems like that > ought to be causing us to skip in this case. Maybe we don't > return early enough? We talked about this on IRC, but for the mailing list: The problem is that piglit_require_GLSL() says "all okay" if you have GL 2.0 or (GL_ARB_shader_objects and GL_ARB_shading_language_100), and R200 exposes both of those extensions, even though it doesn't expose the actual extensions you need for shading... ARB_fragment_shader or ARB_vertex_shader. And moreover, since we support fragment program and vertex program shader_tests, we don't actually want to be requiring GLSL anyway. I'll work on sorting that out, but I think t his patch is good as is (with the GL_ARB_geometry_shader4 change). ___ Piglit mailing list Piglit@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/piglit
Re: [Piglit] [PATCH] ext_framebuffer_blit: Verify framebuffer bindings after calling glGenerateMipmap
On Fri, Nov 13, 2015 at 2:54 PM, Ian Romanickwrote: > From: Ian Romanick > > Mesa's meta path for glGenerateMipmap had a problem that it assumed the > two bindings were the same. As a result, one of them was incorrect > after returning from _mesa_meta_GenerateMipmap. > > NOTE: This currently fails on all Mesa drivers that use > _mesa_meta_GenerateMipmap. > > Signed-off-by: Ian Romanick > --- > tests/all.py | 3 +- > .../spec/arb_framebuffer_object/CMakeLists.gl.txt | 1 + > .../GenerateMipmap-versus-READ_FRAMEBUFFER.c | 129 > + > 3 files changed, 132 insertions(+), 1 deletion(-) > create mode 100644 > tests/spec/arb_framebuffer_object/GenerateMipmap-versus-READ_FRAMEBUFFER.c > > diff --git a/tests/all.py b/tests/all.py > index f520920..431ea30 100644 > --- a/tests/all.py > +++ b/tests/all.py > @@ -2003,7 +2003,7 @@ with profile.group_manager( > g(['fbo-generatemipmap-1d']) > g(['fbo-generatemipmap-1d', 'RGB9_E5']) > g(['fbo-generatemipmap-3d']) > -g(['fbo-generatemipmap-3d', 'RGB9_E5']) > + oops? ___ Piglit mailing list Piglit@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/piglit
Re: [Piglit] [PATCH v1] Porting the OpenGL shading language from Glean to Piglit.
On Tue, Oct 20, 2015 at 12:16 AM, Juliet Fruwrote: > Hi Brian, > > I have been working on the "gl_Position not written check" > (tests/spec/glsl-1.10/linker/) test from glean to shader_test files. Here is > what I have so far. I'm sure there may be some errors since this test does > not have the fragment shader aspects. I am not very familiar with > shader_test files. I will appreciate if you could correct this sample test > so I could submit the patch and move on to the others. Hi Juliet, I was going to port the fragment/vertex program (tfragprog1.cpp, tvertprog1.cpp) tests in glean to regular shader_test files, but I don't want to collide with anything you're doing. I haven't seen anything from you in more than three weeks, so I think your project might be complete? Can you confirm your status? Thanks, Matt ___ Piglit mailing list Piglit@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/piglit
Re: [Piglit] [PATCH] ext_framebuffer_blit: Verify framebuffer bindings after calling glGenerateMipmap
On 11/13/2015 12:04 PM, Ilia Mirkin wrote: > On Fri, Nov 13, 2015 at 2:54 PM, Ian Romanickwrote: >> From: Ian Romanick >> >> Mesa's meta path for glGenerateMipmap had a problem that it assumed the >> two bindings were the same. As a result, one of them was incorrect >> after returning from _mesa_meta_GenerateMipmap. >> >> NOTE: This currently fails on all Mesa drivers that use >> _mesa_meta_GenerateMipmap. >> >> Signed-off-by: Ian Romanick >> --- >> tests/all.py | 3 +- >> .../spec/arb_framebuffer_object/CMakeLists.gl.txt | 1 + >> .../GenerateMipmap-versus-READ_FRAMEBUFFER.c | 129 >> + >> 3 files changed, 132 insertions(+), 1 deletion(-) >> create mode 100644 >> tests/spec/arb_framebuffer_object/GenerateMipmap-versus-READ_FRAMEBUFFER.c >> >> diff --git a/tests/all.py b/tests/all.py >> index f520920..431ea30 100644 >> --- a/tests/all.py >> +++ b/tests/all.py >> @@ -2003,7 +2003,7 @@ with profile.group_manager( >> g(['fbo-generatemipmap-1d']) >> g(['fbo-generatemipmap-1d', 'RGB9_E5']) >> g(['fbo-generatemipmap-3d']) >> -g(['fbo-generatemipmap-3d', 'RGB9_E5']) >> + > > oops? Oops in deed! Fixed locally. Thanks. ___ Piglit mailing list Piglit@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/piglit