The major differences compared to the NV extensions are:
- support for the sequence index as push constants
- support for draw with count tokens (note that DrawID is zero for
normal draws)
- support for raytracing
- support for IES (only compute is supported for now)
- improved preprocessing support with the state command buffer param
The NV DGC extensions were only enabled for vkd3d-proton and it will
maintain both paths for a while, so they can be replaced by the EXT.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31383>
Using view3dAs2dArray changes the tiling and it's slower (-7.5% in
Silent Hill 2 Remake) than using 3D tiling. The previous implementation
was the best one regarding performance (it's also what RadeonSI does).
Sadly it seems that sampler2DViewOf3D can't really be supported without
that but nobody really needs it apparently.
Also view3dAs2array is incompatible for 2D views of sparse 3D images
because sparse 3D images requires 3D tiling.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11997
This reverts commit f5805bcb8e.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31869>
If blending is disabled or the color write mask is 0, dual-source
blending would be ignored, and this can be simplified a bit.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31681>
When the FS is unknown, this can happen with fast-link GPL or unlinked
ESO, rely on the number of VS/TES outputs which should be a good
approximation of the number of PS inputs.
This fixes a (huge?) performance regression from May 2023 because
for depth-only rendering, the FS is NULL and NGG culling wasn't
considered at all.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31830>
This has been removed few years ago by mistake but it's important for
performance. This is mostly for addrlib to determine tile_swizzle which
is used to make memory access faster with multiple render targets.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31797>
We cannot set the {X,Y}MAX_RIGHT_EXCLUSION bits
if we have a sample location at a pixel boundary.
CTS does not seem to be catching this.
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Co-authored-by: Vitaliy Triang3l Kuzmin <triang3l@yandex.ru>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31839>
RADV needs to adjust this register for user sample locations because
it seems possible to have a sample on the -8 coordinate.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31815>
It's possible to enable NGG culling with ESO if shaders are linked, or
if the VS doesn't need a prolog or if TES is used. This wasn't
supposed to be enabled but I think it worked just by luck because the
user SGPR value was probably zero and NGGC was disabled at draw time.
Found by inspection.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31829>
This is useful in order to correlate shader hashes between RGP and
Fossilize. This is because Fossilize needs to pass the capture
statistics flag for getting shader hashes and the pipeline key won't
match otherwise.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31820>
Setting the surface index is optimal for performance in case of
multiple MRTs because addrlib rotates tiles differently.
But this should be disabled when the image is used for descriptor
buffers capture&replay because the descriptor isn't immutable
(ie. tile_swizzle can be different).
This fixes dEQP-VK.binding_model.descriptor_buffer.capture_replay.*
on some GPUs where tile_swizzle is non-zero.
Fixes: 3b57a35ece ("radv: Enable descriptorBufferCaptureReplay.")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31818>
When only of the depth/stencil aspects is used, RADV dispatches a
compute shader to initialize the HTILE buffer. But dispatching on SDMA
just hangs and the only way to initialize the HTILE buffer is to clear
both aspects using a memory fill operation.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31803>
This is a per-component bit mask (0xf for each output).
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes: 0e0c2574d1 ("radv: Add shader stats for inputs and outputs.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31593>
Looks like this was disabled by mistake. RadeonSI relies on the default
value which is "no force" and PAL only sets this to FORCE_DISABLE when
HTILE is completely disabled using settings.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31690>
The Vulkan spec says:
"If the buffer was created with a non-zero value of
VkBufferOpaqueCaptureAddressCreateInfo::opaqueCaptureAddress or
VkBufferDeviceAddressCreateInfoEXT::deviceAddress, the return
value will be the same address that was returned at capture time."
My interpretation is that you can get the buffer device address before
binding if you passed a non-zero address during buffer creation. The
returned BDA would be similar if a memory object is bound to the
buffer later.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31145>
Spec allows calling VkBufferGetDeviceAddressInfo without binding to memory:
VUID-VkBufferDeviceAddressInfo-buffer-02600
If buffer is non-sparse and was not created with the VK_BUFFER_CREATE_DEVICE_ADDRESS_CAPTURE_REPLAY_BIT flag, then it must be bound completely and contiguously to a single VkDeviceMemory object
Which we can only do by making it sparse unconditionally, which feels very wrong to me for a capture & replay extension as that significantly impacts execution.
Current theory is that this was only intended for the EXT and not the core functionality. As such, let's disable capture using the EXT.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31140>
The IB2 packet is only supported on the graphics queue. To execute DGC
IB on compute, the previous solution was to submit it separately
without any chaining. Though this solution was incomplete because it's
easy to reach the maximum number of IBs per submit when there is a lot
of ExecuteIndirect() calls.
To fix that, the proposed solution is to implement DGC IB chaining when
it's executed on the compute only. The idea is to add a trailer that is
added at the beginning of the DGC IB (to know the offset). This trailer
is used to chain back back the DGC IB to a normal CS, it's patched at
execution time. Patching is fine because it's not allowed to execute
the same DGC IB concurrently and the entire solution relies on that.
When the DGC IB is executed on graphics, the trailer isn't patched and
it only contains NOPs padding. Performance should be mostly similar.
This fixes
dEQP-VK.dgc.nv.compute.misc.execute_many_*_primary_cmd_compute_queue.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30809>
Preprocess now must use the same conditional rendering state as the
execute, so the DGC prepare shader must reset the number of sequences
to generate an empty cmdbuf for compute.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31563>
