Meson 1.7.0 is required to ensure that generated bindings
are suitable for use with the selected Rust compiler.
Bindgen is bumped to 0.71.1 which has more permissive handling of
`--rust-target`, enabling (almost) any valid Rust version to be
passed as a target; `-nightly` and `-beta*` suffixes are an exception
to this.
This behaviour is expected to improve further with later versions of
Meson; upstream fixes have landed that:
- pass a stripped version string to bindgen
- additionally pass an appropriate `--rust-edition` if `rust_std` is
set (or "nightly" if the compiler is `nightly`).
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12680
Bug: https://bugs.gentoo.org/946913
Signed-off-by: Matt Jolly <kangie@gentoo.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33725>
We now depend on a later version of bindgen than is available
in the Fedora repositories. Use `build-bindgen.sh` to install
the specific version that we need.
Signed-off-by: Matt Jolly <kangie@gentoo.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33725>
We already use `install-meson.sh` so that we can select an appropriate
version of Meson. Instead of installing two copies just do it before
we begin building anything in the container.
Signed-off-by: Matt Jolly <kangie@gentoo.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33725>
We need to use a later version of Meson than is currently available
in the Alpine repository. Use the existing `install-meson.sh` to fetch
the appropriate version from PyPI.
This requires that we explicitly install a Ninja impl on Alpine -
Samurai is the preferred impl and by using this we do not need
to prefix `PATH` to access Ninja.
`apk` installed Installed packages are additionally sorted.
Signed-off-by: Matt Jolly <kangie@gentoo.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33725>
So Goldfish doesn't really have /dev/dri/renderN128
in it's system image, but somehow this worked in the
past?
Probably since VirtGpuDevice::getInstance() never
returned a nullptr in the past, and now it does.
Add nullptr checks to fix the crash.
TEST=emulator -feature GuestAngle -gpu lavapipe boots
to UI on main
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35580>
When `VkFramebufferCreateInfo::flags` has
`VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT`,
`VkFramebufferCreateInfo::pAttachments` should be ignored.
Test: cvd create \
--gpu_mode=gfxstream_guest_angle_host_swiftshader \
--gpu_renderer_features="VulkanSnapshots:enabled"
Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35580>
nir_unsigned_upper_bound is good enough that this isn't needed anymore.
No fossil-db changes.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35514>
We only want the atomic bit to be conditional to non sparse.
Also take the opportunity to fix buffer features and report the same
supported atomic formats as images.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: ed77f67e44 ("anv: add emulated 64bit integer storage support")
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35358>
The max dimensions are in units of pixels, not bytes. But the x
coordinate shift is based on aligning the address/offset to 64.
Rework the buffer clear loop to iterate in terms of pixels, but
with the x dimension shift based on converting aligned offset
to pixels.
Fixes OpenCL-CTS test_buffers.
Fixes: dafc4476f7 ("freedreno: Implement fast clear_buffer for Adreno 6xx and 7xx")
Signed-off-by: Rob Clark <rob.clark@oss.qualcomm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35557>
This allows sampling from a buffer as if it was a simple (single
level/layer, linear) 2d image/texture. Add a helper to convert
struct pipe_image_view to an apprpriate fdl_layout and wire it up
in the image view and tex view paths.
Can be tested with cl cts:
./test_image_streams 2D read CL_RG CL_SIGNED_INT16 CL_FILTER_NEAREST CL_ADDRESS_CLAMP_TO_EDGE UNNORMALIZED
(for ex)
Signed-off-by: Rob Clark <rob.clark@oss.qualcomm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35447>
- cmdbuf tracepoint moved to tu_cmd_buffer_begin since we want to
trace internal cmdbuf where we reconstruct renderpass.
- Stomp renderpass start tracepoint if renderpass doesn't end in
current cmdbuf, and create new renderpass tracepoint when rp
is reconstructed.
- Create new trace_rp_drawcalls_start when reconstructing rp
in the internal cmdbuf.
With this there are no crashes in:
dEQP-VK.dynamic_rendering.primary_cmd_buff.*
With TU_DEBUG=gmem,forcebin MESA_GPU_TRACES=print
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34981>
There were two issues with the current implementation:
- We didn't set the offset for sampler and image views
- Image::fill didn't take the parents offset into account
Cc: mesa-stable
Reported-by: Rob Clark <rob.clark@oss.qualcomm.com>
Tested-by: Rob Clark <rob.clark@oss.qualcomm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35537>
IsSupported() returns S_OK only when the HW encoder supports this API.
GetValue() returns the supported block size for CODECAPI_AVEncVideoOutputQPMapBlockSize and CODECAPI_AVEncVideoOutputBitsUsedMapBlockSize.
SetValue() returns E_INVALIDARG if the set value does not match the supported value.
User can call GetValue() to get the supported block size before calling SetValue() to avoid the failure.
Reviewed-by: Pohsiang (John) Hsu <pohhsu@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35554>
