Improve: 91fe0b5629 ("radv: Delete lots of sync code.")
As cnd_monotonic.h are include `util/os_time.h`, radv_debug.c and radv_debug.c needs `util/os_time.h`
So include in these files directly.
The compiling errors are:
```
../src/amd/vulkan/radv_debug.c:707:12: error: implicit declaration of function 'os_localtime' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
timep = os_localtime(&raw_time, &result);
../src/amd/vulkan/radv_device.c:97:11: error: implicit declaration of function 'os_time_get_nano' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
return os_time_get_nano();
^
../../src/amd/vulkan/radv_pipeline.c: In function 'radv_create_shaders':
../../src/amd/vulkan/radv_pipeline.c:4119:29: error: implicit declaration of function 'os_time_get_nano' [-Werror=implicit-function-declaration]
4119 | int64_t pipeline_start = os_time_get_nano();
```
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16536>
The callback allows to request different vectorization factors
per instruction depending on e.g. bitsize or opcode.
This patch also removes using the vectorize_vec2_16bit option
from nir_opt_vectorize().
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13080>
DiscardRectangleCount must be less than or equal to
maxDiscardRectangles which is limited to 4.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16778>
The value must be less than or equal to maxTessellationPatchSize which
is limited to 32 with RADV.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16778>
Instead of gathering XFB info over and over again every time we even
want to know if the shader uses XFB, gather it once when we compile the
shader and refer to the copy in the NIR.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
Unfortunately, radv_lower_multiview is not suitable for mesh shaders
because it can't know the mapping between API mesh shader
invocations and output primitives.
Additionally, when lowering view id to layer, it must be created
as a per-primitive PS input.
Fixes: d32656bc65
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16504>
Use vk_buffer as a base for radv_buffer and
replace manual handling of VK_WHOLE_SIZE with
vk_buffer_range.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16764>
This will allow to translate the function and factors earlier.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16672>
There is also a hypothetical scenario where
transformData is 0 and transformOffset is not 0
and we end up reading from transformOffset because
transform_addr is not 0.
VkAccelerationStructureBuildRangeInfoKHR spec:
If VkAccelerationStructureGeometryTrianglesDataKHR::transformData is not NULL, a single VkTransformMatrixKHR structure is consumed from VkAccelerationStructureGeometryTrianglesDataKHR::transformData, at an offset of transformOffset. This matrix describes a transformation from the space in which the vertices for all triangles in this geometry are described to the space in which the acceleration structure is defined.
Which I think means, that we should ignore
transformOffset if transformData is NULL.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16719>
VkAccelerationStructureBuildRangeInfoKHR spec:
If the geometry uses indices, primitiveCount × 3 indices are consumed from VkAccelerationStructureGeometryTrianglesDataKHR::indexData, starting at an offset of primitiveOffset. The value of firstVertex is added to the index values before fetching vertices.
If the geometry does not use indices, primitiveCount × 3 vertices are consumed from VkAccelerationStructureGeometryTrianglesDataKHR::vertexData, starting at an offset of primitiveOffset + VkAccelerationStructureGeometryTrianglesDataKHR::vertexStride × firstVertex.
Meaning: We always add firstVertex * vertexStride
to the vertex address and add primitiveOffset
either to the vertex address or the index address,
depending on wether indices are used.
Also add missing handling with instances.
Fixes: 0dad88b ("radv: Implement device-side BVH building.")
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16719>
We accidentally compared the stencil layout to the color/depth layout.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16663>
The VRS image no longer uses HTILE like on GFX10.3 and I don't know
yet how to program it.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16557>
No FMASK on GFX11 which means I don't think we can expose this ext.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16557>
GFX11 doesn't support CB_RESOLVE which means it doesn't support the
hardware resolve path. If necessary (for arrays or integer formats)
the driver will select the compute path.
No CTS regressions by forcing the fragment resolve path on GFX10.3
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16557>
Found these by diffing the list of registers between GFX10_3 and GFX11.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16557>
According to the Vulkan spec 21.4 "Conditional Rendering",
only clearing attachments with vkCmdClearAttachments is subject to
conditional rendering.
Subpass clear and vkCmdClearColorImage / vkCmdClearDepthStencilImage
should always be executed even if it happens in a
conditional rendering block.
Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16654>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Martin Roukala <martin.roukala@mupuf.org>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16702>
MEC (the compute queue firmware) does not support real
predication, so we have to emulate that using COND_EXEC
packets before each dispatch.
Additionally, COND_EXEC doesn't have an inverted mode, so
in order to support inverted mode conditional rendering, we
allocate a new piece of memory in which we invert the condition.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6533
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16653>
Rename radv_emit_global_shader_pointers to radv_emit_graphics_shader_pointers
and don't call it on compute queues. Note that the registers
that this function sets, are not available on the async compute
queue anyway.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16395>
In the future we'll need a 1:N mapping between radv_queue objects and
HW queues, meaning that 1 radv_queue object will need to be able to
submit to multiple queues. To do that, we'll must also maintain a different
state for each HW queue.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16395>
This will be needed for a future refactor of these functions
where we won't pass radv_queue to them anymore.
Also remove the radv_queue pointer from functions which
don't actually need it.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16395>
This is to prevent excessive code repetition, and hopefully make the
code a bit easier to read.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16395>
Put the shader ring information into a separate structure.
Also clean up how this information is used when generating the preambles
and add a few comments that explain how it works.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16395>
Ray queries and acceleration structure builds
are quite stable now and so we can enable those
features for CI and more feedback and bug reports.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16007>
We use magic numbers for node types in many
places of the RADV code. This commit adds an enum
for the node types and uses it to replace magic
numbers in the acceleration structure build code.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16650>
I should have introduced radv_graphics_pipeline_info earlier. More
states will be added later.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16643>
Instead of using a union in radv_pipeline, this introduces new
structures for graphics, compute and library pipelines which inherit
from radv_pipeline. This will ease graphics pipeline libary implem.
There is still no radv_raytracing_pipeline because RADV actually
uses a compute pipeline for everything but it could be introduced
later when necessary.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16603>
This removes a TON of pipeline->device occurences which were a mess.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16603>
These functions should only emit registers and not change the pipeline.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16603>
This would be needed by some NIR pass during linking. Given that NGG
settings are currently dispatched in many places, I don't think this
should hurt but it should be refactored at some point.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16404>
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Martin Roukala <martin.roukala@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16604>
Except the viewport/scissor arrays to avoid useless array copies.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16552>
It's part of the pre-rasterization state of graphics pipeline libs.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16552>
LLVM uses an implicit argument for the ring offsets and this lowering
was just broken.
This fixes tessellation and geometry on all generations with LLVM.
Fixes: 896a55f47d ("radv: Lower ABI in NIR for tess/ESGS/NGG shader arguments.")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16530>
We do not need this any longer since
radv_acceleration_structure.c uses
radv_fill_buffer now instead.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16517>
It turns out, that the fuchsia sort may actually
perform clears with size < 16 which hits an assert
in radv_fill_buffer_shader. This fixes random
crashes in Control.
Fixes: be57b08 ("radv: Build accaleration structures using LBVH")
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16517>
Makes the bo parameter optional which is useful
for the clears performed by acceleration structure
build commands.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16517>
We can't use UINT16_ABGR for the alpha channel. Always use 32_ABGR.
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16509>
Fixes: 5d9ef0efb5 ("radv: Add the fuchsia radix sort")
../mesa/src/amd/vulkan/radix_sort/radv_radix_sort.c(148): error C2220: the following warning is treated as an error
../mesa/src/amd/vulkan/radix_sort/radv_radix_sort.c(148): warning C4098: 'vkDestroyPipeline': 'void' function returning a value
[82/1129] Compiling C++ object src/amd/compiler/libaco.a.p/aco_instruction_selection.cpp.obj
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16373>
I forgot to use task_ring_offsets instead of ring_offsets
when I ported this code to the new ABI.
Fixes: a8bdcf3c92
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16500>
This is in the wrong place thanks to a rebase mistake.
Fixes: 101a7321c4
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16500>
This pass new dEQP-VK.pipeline.*.image_2d_view_3d_image.*.
This should also fix a GPU hang with Deus Ex: Mankind Divided and Zink.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16294>
The hardware can't bind a slice of a 3D image as a 2D image, GFX10+
introduced ARRAY_PITCH to allow this without a shader workaround.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16294>
Previously this check was skipped for pools with
VK_DESCRIPTOR_POOL_CREATE_FREE_DESCRIPTOR_SET_BIT unset, but after
96a240e1 we need to check this otherwise we risk overflowing
radv_descriptor_pool::entries into the host memory base
This fixes a crash to desktop when launching Dota 2, which overallocates
descriptor sets and expects an error to allocate another descriptor pool
Fixes: 96a240e176 ("radv: fix memory leak of descriptor set layout")
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16490>
From the Vulkan spec:
"VUID-VkGraphicsPipelineCreateInfo-PrimitiveId-06264
If the pipeline is being created with pre-rasterization shader
state, it includes a mesh shader and the fragment shader code
reads from an input variable that is decorated with PrimitiveId,
then the mesh shader code must write to a matching output variable,
decorated with PrimitiveId, in all execution paths"
So, if PS uses PrimitiveID, MS must export it (like GS).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16438>
This merges them into si_llvm_load_intrinsic and reuses load_tess_varyings.
RADV only implemented 1 callback.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16467>
We set the number of task shader ring entries in radv_device
based on the generous assumption that each CU can run task/mesh
shaders with maximum occupancy.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14929>
Mostly the same as for compute shaders, but with a few extras:
task_ring_offsets:
Same as what ring_offsets is to graphics shaders.
Contains an address that points to a buffer that contains
the ring buffer descriptors.
task_ring_entry:
Index that can be used to address the draw and payload rings.
draw_id:
Same meaning as in graphics shaders.
task_ib_addr/task_ib_stride:
Indirect buffer address and stride from the draw calls.
These are used to emulate the firstTask feature of NV_mesh_shader.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14929>
This should be kept to only things aco uses, and expanded when
radeonsi support is added. Things should be removed if lowered in NIR.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16342>
Avoid unnessecary work on FOREIGN queue release barriers. If we can't modify
the image there can't be a situation where we need to update the presentable
dcc data.
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16371>
This pattern pops up a bunch and the semantics of nir_channels() aren't
very convenient much of the time. Let's add a nir_trim_vector() which
matches nir_pad_vector().
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16309>
nir_texop_samples_identical is lowered to
nir_texop_fragment_mask_fetch_amd earlier.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16426>
This takes the buffer ptrs out of the struct, so they can be memset
separately and optimises the memset to be minimal for them.
Removing them from the struct avoids having to loop to clear them.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16393>
This code path will be used on kernels that don't have the
"gang submit" feature in amdgpu yet.
It does each submission separately, and uses scheduled dependencies
to connect them.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16361>
This prepares RADV for submitting to multiple queues at the same time.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16361>
This makes the function easier to read and follow.
Also prepares for some future changes when we'll want to
submit to multiple queues at once.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Instead, just use the pointers from the queue structure.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
This function already handles more than just the geometry and tess
rings, and it will include more things in the future, such as
the task shader rings.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Previously we would always create it, but would return NULL when
it wasn't needed. Now, don't create it when not needed.
Additionally, don't create the continue preamble when we can use
IB BOs, because then we never use a continue preamble.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
The "initial" and "initial full flush" preambles both need a
cache flush with a slight difference. Improve the readability of
this code.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
This aims to improve the readability of this function.
Initialize the BO pointers to the queue's current pointers
at the beginning instead of in the else branches later.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Hopefully this will make it easier to read and understand.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Also move the preamble update into this function, as that is only needed
by this code path and not needed for empty submits.
With this change, the goal is to make radv_queue_submit easier to
read and understand. This prepares it for future work when we'll
add the capability to submit to multiple queues at the same time.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16358>
Additions by positive integers can more easily be combined into the
access.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
Use integer atomics on GFX8/9.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16203>
RDNA2 is affected by a hardware bug when instance packing is enabled
for adjacent primitive topologies and instance_count > 1, pipeline
stats generated by GE are incorrect. It needs to be applied for
indexed and non-indexed draws.
This is based on PAL waDisableInstancePacking.
This fixes KHR-GL46.pipeline_statistics_query_tests_ARB.* with Zink.
Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6257
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15877>
This is to fix a LLVM crash with 13.0 because ATOMIC_FMAX is only
supported on 14.0+, so RADV_DEBUG=llvm was just completely broken.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16305>
This test can intentionally make the alloc callback to return NULL, so
we have to handle object creation failures properly. The driver should
also avoid memleaks because the test checks that.
Fixes crashes with
dEQP-VK.api.device_init.create_instance_device_intentional_alloc_fail.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Martin Roukala <martin.roukala@mupuf.org>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16298>
Fixes dEQP-VK.binding_model.descriptorset_random.sets4.noarray.ubolimitlow.sbolimitlow.sampledimglow.outimgtexlow.noiub.nouab.rgen.noia.0
and probably some other ones.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16306>
The dynamic primitive topology could change from LINE_LIST to
LINE_STRIP for example and the stipple state depends on this.
Found by inspection.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16295>
Without --quiet glslang unconditionally prints the input file name to stdout.
Check if --quiet is supported because some distros only have ancient glslang
versions.
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16312>
Instead of binding 2D images views and relying on BASE_ARRAY.
This cleanups the internal driver operations for 3D.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16272>
We usually use pdevice for "physical device" and not "device pointer".
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16259>
With NGG GS, the hardware can't know the number of generated primitives
and we have to increment it manually from the shader using a plain GDS
atomic operation.
Though this had a serious problem (see this old TODO) if the bound
pipeline was using legacy GS because the query implementation was
relying on NGG GS. Another situation is if we had one draw with NGG GS,
followed by one draw with legacy (or the opposite) the query result
would have been broken.
The solution is to allocate two 64-bit values for storing the begin/end
values if the query pool is supposed to need GDS and accumulate the
result with the number of generated primitives generated by the hw.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15892>
This controls the whole lowering of "make tex ops with implicit
derivatives on non-implicit-derivative stages be tex ops with an explicit
lod of 0 instead", but it's really hard to describe that in a git commit
summary.
All existing callers get it added except:
- nir_to_tgsi which didn't want it.
- nouveau, which didn't want it (fixes regressions in shadowcube and
shadow2darray with NIR, since the shading languages don't expose txl of
those sampler types and thus it's not supported in HW)
- optional lowering passes in mesa/st (lower_rect, YUV lowering, etc)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16156>
To workaround game bugs where partial derivatives are used in
non-uniform control flow. A proper solution needs to be implemented,
but as a quick fix disabling nir_opt_sink() works.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16165>
When the divisor is 0, the compiler should generate a different VS
prolog instead of re-using a previous prolog that uses nontrivial
divisors. This is because divisor == 0 and divisor > 1 should use
a different path to guarantee that the index is correctly computed.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16009>
The vkd3d-proton ray tracing tests delete shader modules after creating
pipeline libraries from them. This resulted in a use after free when
creating ray tracing pipelines.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16027>
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: 15640e58d9 ("radv,aco: lower texture descriptor loads in NIR")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16148>
Fixes dEQP-VK.ray_tracing_pipeline.misc.report_intersection_result
Fixes: c3d82a9 ("radv: Add pass to lower anyhit shader into an intersection shader.")
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16056>
To clarify that other clear operations like vkCmdFillBuffer() are
implicitly synchronized. So, STAGE_2_CLEAR_BIT is only needed for
vkCmdUpdateBuffer() in some rare cases (GFX10+ dGPUs with GTT BOs).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15943>
This is the result of running clang-format on the entire file.
Although this is not related to this MR, I decided to include
the formatting changes so I do not have to deal with them if
I work on this file in the future.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15648>
There was a bug that inst_offset was added to soffset in one codepath and
to voffset in all other codepaths. The correct behavior is to add it
to voffset.
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15966>
Getting opengl32*.def consistence with Windows SDK.
Getting osmesa.mingw.def's gl* functions consistence with Windows SDK.
stw_* functions are cdecl, not stdcall, so there is no need mangling the symbol.
Fixes egl.def for x86
d3d10sw: Move the place of d3d10_sw.def to d3d10_sw.def.in
Fixes vulkan_lvp.def for x86
Fixes#5552
Remove stdcall-fixup
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14041>
Looks like it can't be supported. Also disabled by PRO/AMDVLK.
Fixes new CTS dEQP-VK.texture.texel_buffer.uniform.srgb.*.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16010>
Instead of copying states from these structures at many different
places, do it only once. Will help VK_EXT_graphics_pipeline_library.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15967>