summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
8 daysarb_fragment_layer_viewport: don't test out-of-range gl_Layer valuesHEADmainIván Briano3-0/+9
Tests try drawing into layer 1 of an framebuffer with no layers, so it's still out of range and thus invalid. Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/913>
12 daysglsl-1.30: Test uint to float conversion of sign extended valueIan Romanick2-0/+90
Intel compiler has a very old optimization for various combination of u2f or i2f of an extract_* source. However, the variations that do u2f of an extract_i* are incorrect. Intel drivers with the known bug will produce blue. v2: Use 0x000000ff in the int8 test to match the int16 test. Suggested by Caio. v3: Update header comment to explain why two possible incorrect values are checked. Suggested by Caio. Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/883>
2024-04-26arb_draw_buffers_blend: Fix deprecated-non-prototype warningsVinson Lee1-2/+2
tests/spec/arb_draw_buffers_blend/state_set_get.c:91:12: warning: passing arguments to a function without a prototype is deprecated in all versions of C and is not supported in C23 [-Wdeprecated-non-prototype] 91 | glEndList(list); | ^ tests/spec/arb_draw_buffers_blend/state_set_get.c:136:12: warning: passing arguments to a function without a prototype is deprecated in all versions of C and is not supported in C23 [-Wdeprecated-non-prototype] 136 | glEndList(list); | ^ Fixes: 0fb057c06ddc ("Add new arb_draw_buffers_blend-state_set_get test") Signed-off-by: Vinson Lee <vlee@freedesktop.org> Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/906>
2024-04-26arb_sample_shading: do not actually run mask_in_one if it's not supportedConstantine Shablia1-0/+3
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/907>
2024-04-24clear-stencil: new test to check for problems clearing stencilsEric R. Smith2-0/+232
On some architectures clearing depth and stencils can be optimized, but tracking the state to know when the stencil has been written is necessary and sometimes has bugs. This test finds a problem with state tracking of stencil clears in an older version of the panfrost driver. Signed-off-by: Eric R. Smith <eric.smith@collabora.com> Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/905>
2024-04-24read_depth_offset: new test to check depth offset calculationEric R. Smith2-0/+253
We have tests to validate depth offsets in general, but these apply to the default depth buffer (usually 24 bit fixed point). This test checks the details of the depth offset calculation by using NV_read_pixel to read back the written values, and verifies that e.g. the minimum resolvable difference used for floating point buffers matches the spec. This revealed a bug on panfrost. Signed-off-by: Eric R. Smith <eric.smith@collabora.com> Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/900>
2024-04-22replayer: create bin/ everytimeDavid Heidelberg1-0/+2
When we build only a small subset of features, the bin directory isn't prepared yet, so we need to be sure, it's there for us. Reviewed-by: Vignesh Raman <vignesh.raman@collabora.com> Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/903>
2024-04-19Reduce precision in depth clear precision checkJuan A. Suarez Romero1-0/+5
For depth16 and depth24, let's reduce the precision of the float values used as depth. This makes the depth-clear-precision tests pass on rpi4 devices. Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com> Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/899>
2024-04-19Add float from half functionJuan A. Suarez Romero2-1/+36
Converts a 16-bit float to 32-bit float. Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com> Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/899>
2024-04-17tests/gbm: Add gbm_device_get_fd() testDaniel Stone2-0/+54
Make sure that GBM doesn't dup() the FD from under us, as users rely on this being the case. Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/865>
2024-04-16wayland: Add Wayland dmabuf driver-name testDaniel Stone8-1/+317
This test checks to see whether the Wayland compositor advertises the expected device as its primary device through the dmabuf interface, so can be used as a basic smoke test for checking that clients are able to use the right renderer. Signed-off-by: Daniel Stone <daniels@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/796>
2024-04-15conversion-explicit: use a different value for normalized +/- minMike Blumenkrantz1-2/+2
this should yield correct results on all drivers fixes mesa/mesa#10991 Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/904>
2024-04-12cmake: if we specify PIGLIT_BUILD_GL_TESTS OPENGL_FOUND is always trueDavid Heidelberg1-1/+1
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12README: recommend Ninja by default and switch to cmake --buildDavid Heidelberg1-10/+6
- Using make is slow. Use Ninja where possible. - Use cmake --build, since depending on cmake settings it'll utilize the right buildsystem. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: build is compatible up to 3.28David Heidelberg1-1/+1
Removes annoying warning. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: add workaround for CMP0148David Heidelberg1-0/+3
This is needed for following modern cmake versions. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12README: ccmake is not that common, regular cmake is enough to buildDavid Heidelberg1-1/+1
For ccmake is necessary to install cmake-curses-gui which isn't a common package. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12README: update requirementsDavid Heidelberg1-1/+2
- We don't need X11 libraries for EGL only build. - User may want use Ninja for faster build. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: WGL doesn't need Waffle, pass the definition without itDavid Heidelberg1-1/+1
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: remove this 12 years old workaround taking too many linesDavid Heidelberg1-14/+0
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: implement switch for the EGL tests and simplify test conditionsDavid Heidelberg1-31/+15
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: make PIGLIT_BUILD_DMA_BUF_TESTS aligned with GBM dependencyDavid Heidelberg1-18/+17
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: depend on WGL_TESTS being enabled to build WGLDavid Heidelberg1-5/+5
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: implement GBM and Wayland controlsDavid Heidelberg1-29/+46
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: fix unusually formated statementDavid Heidelberg1-2/+2
Not very common for if,else,endif. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: implement X11 option (required for GLX)David Heidelberg1-7/+14
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: move GLX and WGL option to the beggining of the fileDavid Heidelberg1-13/+3
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: move first batch of option() at the beggining of the fileDavid Heidelberg1-11/+11
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: rename default values with prefix DEFAULT_David Heidelberg1-10/+10
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: move defaults into the per-platform sectionDavid Heidelberg1-8/+6
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: predefine defaults depending on the systemDavid Heidelberg1-5/+26
Make CMakeLists.txt cleaner. Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: drop unused CMP0042 policyDavid Heidelberg1-5/+0
Currently required cmake 3.2 defaults to NEW (e.g. ON). Fixes: fdc5a4ca1112 ("cmake: Bump minimum version to 3.2.") Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-12cmake: respect indentationDavid Heidelberg1-16/+14
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/898>
2024-04-10texturing: add a test for GPU/CPU sync when creating texturesEric R. Smith3-0/+104
On Panfrost, at least, a race condition could occur if a texture is cleared and then initialized with texSubImage; the clear on the GPU could happen after the memcpy in texSubImage. This test checks for that and triggers the original Panfrost bug. Signed-off-by: Eric R. Smith <eric.smith@collabora.com> Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/901>
2024-03-26util: fix multisample_texture_fs_templateErik Faye-Lund1-2/+2
It's not the texture function that needs the prefix, it's the sampler. Fixes: c55e22a5c ("arb_copy_image-formats: avoid undefined behavior") Reviewed-by: Eric R. Smith <eric.smith@collabora.com> Tested-by: Eric R. Smith <eric.smith@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/897>
2024-03-26tests: add a test for GL_APPLE_syncLucas Stach13-0/+714
Tests from desktop GL ARB_sync adapted to GLES extension APPLE_sync. Signed-off-by: Lucas Stach <dev@lynxeye.de> Reviewed-by: Adam Jackson <ajax@redhat.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/844>
2024-03-25ci: codespell ignore list is lowercaseLucas Stach1-1/+1
The somewhat twisted language in [1] seems to mean that words in the ignore list must be passed as lowercase. Discussion in [2] supports this understanding. Change all matches to lowercase and remove uppercase entries that are already matched by lowercase ones. [1] https://github.com/codespell-project/codespell/#ignoring-words [2] https://github.com/codespell-project/codespell/issues/2375 Signed-off-by: Lucas Stach <l.stach@pengutronix.de> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/894>
2024-03-25more spelling fixes accross the whole codebaseLucas Stach30-42/+42
While c10a0731c940 ("fix the spelling in whole piglit") already fixed a multitude of spelling issues, new versions of codespell find quite a bit more worth fixing. Signed-off-by: Lucas Stach <l.stach@pengutronix.de> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/895>
2024-03-25ci: bump distribution tagLucas Stach1-1/+1
Needed for 9405c6e80387 ("ci: teach codespell to accept my name") to to take effect. Signed-off-by: Lucas Stach <l.stach@pengutronix.de> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/893>
2024-03-21arb_multisample: Fix bitwise-instead-of-logical warningVinson Lee1-1/+1
tests/spec/arb_multisample/pushpop.c:144:9: warning: use of bitwise '&' with boolean operands [-Wbitwise-instead-of-logical] 144 | return test_enable_bits(enable_on) & test_coverage(coverage_mode); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | && tests/spec/arb_multisample/pushpop.c:144:9: note: cast one or both operands to int to silence this warning Fixes: 1f58a74ca ("ARB_multisample/pushpop: New test for PushAttrib/PopAttrib.") Signed-off-by: Vinson Lee <vlee@freedesktop.org> Reviewed-by: Lucas Stach <l.stach@pengutronix.de> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/857>
2024-03-20ext_texture_storage: Fix memory leak on error pathVinson Lee1-1/+4
Fix defects reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable name going out of scope leaks the storage it points to. leaked_storage: Variable opt going out of scope leaks the storage it points to. Fixes: cc1e4d6761a9 ("Add GL_EXT_texture_storage tests") Signed-off-by: Vinson Lee <vlee@freedesktop.org> Reviewed-by: Lucas Stach <l.stach@pengutronix.de> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/879>
2024-03-20arb_internalformat_query2: fix internalformat selection for max_layers subtestLucas Stach1-1/+1
The wrong index variable used leads to effectively random results when the GPU doesn't support all of the tested internalformats. Signed-off-by: Lucas Stach <l.stach@pengutronix.de> Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/889>
2024-03-20arb_clear_texture: only test float depth textures when supportedLucas Stach1-2/+10
Don't test clearing float depth textures when the driver doesn't advertise support for them. Signed-off-by: Lucas Stach <l.stach@pengutronix.de> Reviewed-by: David Heidelberg <david.heidelberg@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/890>
2024-03-20properly check for libgen.hErik Faye-Lund4-3/+8
Some users are reporting that basename is not available when building on Linux with musl libc. And since the POSIX spec[1] says that basename is defined in libgen.h, we should include that when available. So let's properly detect the header, and include it if it exists. This should hopefully make things a bit more robust. Since we're also including this from the CL program-tester, let's rearrange the includes a bit so we know that config.h has been included. Reviewed-by: David Heidelberg <david.heidelberg@collabora.com> Tested-by: David Heidelberg <david.heidelberg@collabora.com> Reviewed-by: Khem Raj <raj.khem@gmail.com> Tested-by: Khem Raj <raj.khem@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/888>
2024-03-19arb_copy_image-formats: avoid undefined behaviorEric R. Smith2-56/+88
Make piglit_multisample_texture() handle integer formats correctly. From the GL spec: "If the values written by the fragment shader do not match the format(s) of the corresponding color buffer(s), the result is undefined." Therefore we create the fragment shader with the appropriate output format based on the internal format used. Signed-off-by: Eric R. Smith <eric.smith@collabora.com> Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/887>
2024-03-10intel/ppgtt: Use getpagesize fallback with MinGWVinson Lee1-1/+1
Fix MinGW build error. tests/general/ppgtt_memory_alignment.c: In function 'piglit_init': tests/general/ppgtt_memory_alignment.c:99:40: error: implicit declaration of function 'getpagesize' [-Wimplicit-function-declaration] 99 | const unsigned int page_size = getpagesize(); | ^~~~~~~~~~~ Signed-off-by: Vinson Lee <vlee@freedesktop.org> Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/885>
2024-03-09README: add Vulkan into the generic descriptionDavid Heidelberg1-3/+3
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com> Signed-off-by: David Heidelberg <david@ixit.cz> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/866>
2024-03-07fix arb_query_buffer_object-qbo to warn on invalid GS results instead of failMike Blumenkrantz1-3/+20
this behavior is technically valid, and the test is mostly useless Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/884>
2024-03-05tests: Fix cl test ↵jiajia Qian1-1/+2
program@execute@builtin@builtin-float-degrees-1.0.generated issue about tolerance. Initially, there is no tolerance for degrees operations set in gen_cl_common_builtins.py, which will cause the test to fail. According to OpenCL related standards, the error value of the result calculated by the degrees function can be within the range of 2ulp. Modified the tolerance of degrees function in the gen_cl_common_builtins.py file. Signed-off-by: jiajia Qian <jiajia.qian@nxp.com> Part-of: <https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/860>
2024-02-29replayer: add ANGLE backendDavid Heidelberg1-0/+71
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-15 11:58:23 +0000