GitLab

Make it explicit, in case it was redirected for the parent process.

See: https://bugs.python.org/issue44158

Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Martin Peres <martin.peres@mupuf.org>
Part-of: <!534>

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <!527>

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Ported from the corresponding ARB_clear_texture test.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Ported from the corresponding ARB_clear_texture test.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Ported from the corresponding ARB_clear_texture test.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Ported from the corresponding ARB_clear_texture test.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Ported from the corresponding ARB_clear_texture test.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Ported from the corresponding ARB_clear_texture test.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

These tests have been ported from the ARB_clear_texture tests
and combine base formats, float, srgb, and integer formats.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Acked-by: Adam Jackson <ajax@redhat.com>
Part-of: <!519>

Just like with commit a0e847e3

 but this is for the stencil test.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eleni Maria Stea <elene.mst@gmail.com>
Part-of: <!528>

The purpose of this test is to use a Vulkan stencil buffer from OpenGL.
The test renders a quad with Vulkan and sets the stencil value to 1
for every pixel that is rendered. The stencil buffer is then imported in
OpenGL and is used as the depth/stencil attachment of an OpenGL FBO. The
color attachment of the FBO is cleared to green and is used as render
target where a blue quad is rendered with the stencil test enabled and a
stencil mask/reference combination that permits drawing only where the
stencil value is 0. The final result must be a green quad with a smaller
blue quad in the middle.

v2:  fixed commit message
v3:  fix in probe (the colors were reversed accidentally)
v4:  special case for AMD that doesn't support 24/8 depth/stencil
v5:  removed special case and modified the test to run subtests per
format which are skipped if a format is not supported
v6:  removed empty space
v7:  renamed all_formats to depth_stencil_formats that is more
meaningful, removed trailing whitespaces
v8:  changed the loop of stencil subtests to skip the depth-only format
v9:  fixed indentation
v10: added base and max level (TexParameteri) for OpenGL color texture,
     fixes incomplete_tex mesa error (MESA_DEBUG=incomplete_tex)
v11: added the test to opengl.py
v12: removed double depth test state change, and disabled the stencil
right after rendering to its corresponding FBO
v13: added stencilLoadOp, stencilStoreOp values in create_renderpass in
vk.c
v14: removed state change from gl_subtest_init since we re-change it in
run_subtest right afterwards
Signed-off-by: Eleni Maria Stea <estea@igalia.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!508>

Required additions to the Vulkan framework to support depth and
stencil buffer interoperability tests.

v2: removed empty line
v3: setting the stencil OP values only once instead of twice
Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!508>

The barriers we use at the end of renderpass should use dstAccessMasks
that are supported by the pipeline stage. Fixed the function that
selects the access mask. Documentation:
https://www.khronos.org/registry/vulkan/specs/1.2-extensions/html/vkspec.html#synchronization-access-types-supported

Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!526>

This makes piglit-runner behave better, where it doesn't report a skip
as a crash.
Reviewed-by: Emma Anholt <emma@anholt.net>

Driver might not support the format for framebuffer objects and could be
marking framebuffer as incomplete. Skip the format if this is the case.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eleni Maria Stea <elene.mst@gmail.com>
Part-of: <!523>

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!521>

Per the S3 documentation, resource names will be quoted when signing on
the host side. So do the same in the client so the signatures can match.
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Part-of: <mesa/piglit!520>

vk_semaphores test doesn't make any use of the external depth buffer,
and so clearing the depth buffer bit is not required. Removed it.

v2: Fixed compile error (Tapani Pälli)
Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!513>

In vk_clear_color we need to transition both color and depth buffers at
the end of renderpass because the last known color attachment layout is
VK_LAYOUT_COLOR_ATTACHMENT_OPTIMAL, and not VK_LAYOUT_GENERAL we used to
clear the image.
Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!513>

Previously we were inferring it from the URL, but this becomes more
complicated if the download URL isn't that of the MinIO server (eg.
there's a proxy in the middle).
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: Rohan Garg <rohan.garg@collabora.com>
Part-of: <mesa/piglit!509>

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <mesa/piglit!518>

Patch changes cmake to always compile shaders instead of pre-compiled
spirv blobs that are about to be removed.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <!518>

Patch introduces toggle for vulkan tests and checks to require libvulkan
and glslangValidator when toggle is enabled (defaults ON with Linux).

v2: require glslangValidator, not glslc
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <mesa/piglit!518>

Now that the gbm path covers everything Intel does, and is no longer
built by default, delete the code.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <!516>

I needed it for debugging why my probe for new libdrm was failing.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <!516>

We need it to build the gbm path of the dma-buf tests.  The container
already had libgbm.so.1, we just needed the dev package.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <!516>

Like on the intel-specific path, allocate ARGB8888 and the test will
fill in with the AYUV/XYUV pixels.  Fixes failures on freedreno.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <!516>

Fix these warnings with pytest >= 4.5.0.

PytestUnknownMarkWarning: Unknown pytest.mark.raises - is this a typo?  You can register custom marks to avoid this warning - for details, see https://docs.pytest.org/en/latest/mark.html

PytestUnknownMarkWarning: Unknown pytest.mark.timeout - is this a typo?  You can register custom marks to avoid this warning - for details, see https://docs.pytest.org/en/latest/mark.html

PytestUnknownMarkWarning: Unknown pytest.mark.slow - is this a typo?  You can register custom marks to avoid this warning - for details, see https://docs.pytest.org/en/latest/mark.html

Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <mesa/piglit!302>

When trying to build with a lower vulkan version, define VK_NULL_HANDLE
when not present and return NULL instead of VK_NULL_HANDLE as
VK_NULL_HANDLE is a flag not a empty struct.
Signed-off-by: Rohan Garg <rohan.garg@collabora.com>
Reviewed-by: Eleni Maria Stea <elene.mst@gmail.com>
Part-of: <!517>

The Intel OpenGL drivers and gbm are both part of the Mesa project.
Therefore, assuming the OpenGL driver supports the graphics device,
then gbm should support it as well.

The libdrm installed on the system may be older and unable to
recognize the graphics device. This can lead to the dmabuf tests
failing simply due to an outdated libdrm.

For this reason, it is preferable to rely on gbm if it available.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <!472>

Tests that are testing the color attachment behavior shouldn't enable
the depth test and shouldn't require a renderbuffer for depth attachments
in OpenGL side.

Changed the following tests:
- vk-image-overwrite
- vk-semaphores-2
Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!512>

v2: Add GBM support.  Suggested by Tapani.

Part-of: <!491>

v2: Add GBM support.  Suggested by Tapani.

Part-of: <!491>

Future commits will add support for formats with 2 bits of alpha, and
those formats cannot reproduce these values.  Adjust the values to
something that 2 bits of alpha can do.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!491>

Added the semaphore tests of ext_external_objects to opengl.py
Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!511>

Test vk-image-display-overwrite could only overwrite the first frame.
That was caused because we had the depth test enabled while there was no
depth FBO attachment (only RBO). Removed the line that enables the depth
test for the test to render as it should.

v2:
* When depth test is disabled, depth buffer bit shouldn't be cleared (Tapani
  Pälli)
* Changed the middle band's color from gray to white to avoid rounding
  errors.

v3:
* Removed depth attachment from GL framebuffer, we don't need it as the
  test is about color images. (Tapani Pälli)
Signed-off-by: Eleni Maria Stea <elene.mst@gmail.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <!510>

build-debian.sh wasn't using all available CPU cores, build-mingw.sh
was using too many.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <!503>