Alejandro Piñeiro [Sun, 2 Aug 2020 00:23:16 +0000 (02:23 +0200)]
v3dv/formats: fix exposing FEATURE_UNIFORM/STORAGE_TEXEL_BUFFER_BIT
If the formats are not suitable as texture type, then they can't be
used as texel buffers.
Gets tests like the following one:
dEQP-VK.image.load_store.without_format.buffer.r32g32b32_sfloat_minalign_uniform
to be properly skipped (instead of Crash on the simulator)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 31 Jul 2020 16:25:06 +0000 (18:25 +0200)]
v3dv: handle multisample image clears
Fixes:
dEQP-VK.pipeline.framebuffer_attachment.*_ms
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 31 Jul 2020 12:58:42 +0000 (14:58 +0200)]
v3dv: handle multisample resolves for formats that don't support TLB resolves
The TLB multisample resolve feature is only limited to specific format types.
For everything else, including sfloat and integer formats, we need to
fallback to a blit resolve. This needs to be handled both for in-pass
resolves as well as for vkCmdResolveImage.
Because these blits would happen after the tile store operations, we need
to make sure we store the multisampled buffers so we can then read them for
the blit resolve.
Fixes the remaining test failures in:
dEQP-VK.renderpass.suballocation.multisample_resolve.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 31 Jul 2020 12:59:09 +0000 (14:59 +0200)]
v3dv: handle multisample resolve of integer formats
The multisample resolve of an integer framebuffer should just take one
of the samples instead of averaging.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 31 Jul 2020 12:52:58 +0000 (14:52 +0200)]
v3dv: fix blitting of signed integer formats
For these we want to select a signed integer output format
and a signed sampler type.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 31 Jul 2020 12:50:57 +0000 (14:50 +0200)]
nir/glsl: add a glsl_ivec4_type() helper
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 31 Jul 2020 07:26:06 +0000 (09:26 +0200)]
v3dv: amend tile size tables with smallest tile sizes available
We'll need this for some cases involving maximum number of multisampled
color targets.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Thu, 30 Jul 2020 12:35:43 +0000 (14:35 +0200)]
v3dv/device: fix minTexelBufferOffsetAlingment
As we understand that texture accesses should be aligned to the UIF
block size.
Fixes several of the CTS tests under this pattern:
dEQP-VK.binding_model.shader_access.primary_cmd_buf.uniform_texel_buffer.*.offset_nonzero
dEQP-VK.binding_model.shader_access.primary_cmd_buf.storage_texel_buffer.*.offset_nonzero
Note: for those tests, using a lower value (64) was enough to get them
working, but again, we understand that the real alignment is the UIF
block size.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Thu, 30 Jul 2020 12:32:26 +0000 (14:32 +0200)]
v3dv: add v3dv_limits file
There are several definitions for hw limits on v3dv_image that we want
to share, but v3dv_private was already growing bigger and messier.
So let's move them to a specific header. Note that there is already a
broadcom/common/v3d_limits.h. We are not putting them there because
right now they are only used by the Vulkan driver, but are candidates
to be moved.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 28 Jul 2020 22:28:28 +0000 (00:28 +0200)]
v3dv/descriptor: support for UNIFORM/STORAGE_TEXEL_BUFFER
This gets passing most uniform/storage_texel buffer tests.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 30 Jul 2020 09:29:12 +0000 (11:29 +0200)]
broadcom/compiler: handle gl_SampleMask writes in fragment shaders
We didn't need this until now, since this was included with GLES 3.2,
but we need it for Vulkan.
Eric had already done the plumbing for it though, we just need to
actually emit the mask.
Fixes some tests in:
dEQP-VK.renderpass.suballocation.multisample_resolve.*
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 30 Jul 2020 07:00:07 +0000 (09:00 +0200)]
v3dv: handle multisampled image copies with the blit path
This should be able to handle partial copies of multisampled images.
This change extends our blit shader interface to also handle multisampled
destinations so that if the blit destination is a multisampled image,
the blit will rely on sample rate shading to copy all samples from
the source image (which must have a matching number of samples).
I have not found any tests in CTS that do partial copies of
multisampled images, so I tested this with a full multisampled image
copy, using this test:
dEQP-VK.api.copy_and_blit.core.resolve_image.whole_copy_before_resolving.4_bit
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 29 Jul 2020 07:49:47 +0000 (09:49 +0200)]
v3dv: add a blit fallback path for vkCmdResolveImage
This fallback is required when we have to do partial resolves. It
works the same way as other blit fallbacks for copy operations: it
will bind the source image as a source texture and blit the selected
region to the destination image.
The difference in this case is that the source image is multisampled
and the blit shader needs to fetch and average individual samples for
each texel.
This gets us to pass all the remaining test cases in
dEQP-VK.api.copy_and_blit.core.resolve_image.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 29 Jul 2020 07:49:08 +0000 (09:49 +0200)]
v3dv: setup texture shader state correctly for multisampled images
Fixes multisampled cases in:
dEQP-VK.pipeline.multisample.sampled_image.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 28 Jul 2020 08:33:17 +0000 (10:33 +0200)]
v3dv: handle multisampled image copies in the TLB path
vkCmdCopyImage can be used to copy multisampled images. We can
easily support that on the TLB path, which copies full images.
For partial copies we will need to amend our blit shader path
to support multisampling resolve.
Fixes:
dEQP-VK.api.copy_and_blit.core.resolve_image.whole_copy_before_resolving.4_bit
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 28 Jul 2020 08:50:01 +0000 (10:50 +0200)]
v3dv: implement vkCmdResolveImage for whole images
For partial resolves we will need a shader blit & resolve fallback.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 28 Jul 2020 06:53:19 +0000 (08:53 +0200)]
v3dv: implement subpass multisample rendering and resolve
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 24 Jul 2020 08:42:43 +0000 (10:42 +0200)]
v3dv: process VkPipelineMultisampleStateCreateInfo properly
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 24 Jul 2020 08:33:16 +0000 (10:33 +0200)]
v3dv: consider MSAA when computing frame tiling
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Sun, 26 Jul 2020 23:51:36 +0000 (01:51 +0200)]
v3dv/pipeline: handle properly OUT_OF_HOST_MEMORY error when allocating p_stage
So far we were just assuming that it would work (so we could try to
access a NULL pointer), and not freeing it properly.
Fixing that was somewhat messy due pipeline_compile_graphics using a
temporary array and then update the internal pipeline stages. As we
are here we just removed the array and simplified
pipeline_compile_graphics code.
Fixes following tests:
dEQP-VK.api.object_management.alloc_callback_fail.graphics_pipeline
dEQP-VK.api.object_management.alloc_callback_fail_multiple.graphics_pipeline
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Sat, 25 Jul 2020 10:44:22 +0000 (12:44 +0200)]
v3dv/pipeline: pre-generate more that one shader variant
In order to reduce the number of shader builds after pipeline creation
(that ideally shouldn't happen) we pre-generate two shader variants at
pipeline creation time. In addition to the default one, that set the
return size for all texture to 16 bit, we build another variant
setting the return size for all textures to 32-bit. cmd buffer selects
the latter if any of the textures requires 32bit.
So we are using an all 16-bit return size or an all 32-bit return size
variants. This could be slightly improved by pre-generating return
size combinations if the texture number is below a threshold. But that
would require more space, and bigger pipeline creation time, so would
need to be evaluated.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Fri, 24 Jul 2020 22:11:51 +0000 (00:11 +0200)]
v3dv/pipeline: when looking for a variant, check first current variant
So far, when checking for a variant fulfilling a specific v3d key, we
were checking the caches, and if that failed, we compiled a new
variant, and update the current variant.
But we could check first if the current variant fullfils that. This
was not really problematic so far, as checking on the caches was fast,
but now that we could be without any kind of shader cache using
V3DV_ENABLE_PIPELINE_CACHE, it is far better to check first current
variant.
Without this vkQuake3 at 720p drops to 1fps when disabling the cache.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Fri, 24 Jul 2020 13:47:23 +0000 (15:47 +0200)]
v3dv/pipeline: remove custom variant cache
Now that we have a default pipeline cache, we can rely on it. This
allows to remove some code, and avoid the need to have a cache per
each pipeline stage.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Wed, 22 Jul 2020 00:08:06 +0000 (02:08 +0200)]
v3dv/pipeline_cache: add default pipeline cache
That it would be used as fallback. Three advantages:
* Having a cache for user operations even if the user doesn't
provide it.
* Having a cache for internal operations. v3dv_meta_copy creates
pipelines for some copy path, so it is interesting to have them
cached.
* Testing: so now the pipeline cache is tested by more CTS tests.
As any other pipeline cache, it can be disabled with the
V3DV_ENABLE_PIPELINE_CACHE. It was suggested that would make sense to
have a specific envvar for the default pipeline cache, but for now
just one envvar is enough.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Thu, 23 Jul 2020 09:18:11 +0000 (11:18 +0200)]
v3dv/pipeline: provide a shader_sha1 to private ShaderModules
So far for private pipelines we were creating dummy shader modules
where we directly provided the nir shader. But for the pipeline cache
we were using the SPIR-V to generate part of the cache key sha1.
The main use case for private pipelines are meta_copy/clear. Those nir
shaders depend on parameters like the format etc, so we use directly
the serialized form of the NIR shader to generate the sha1.
The other case are the no-op fragment shader that we need to provide
if no fragment shader is defined by the user. For that case we can
just use the default shader name, as the no-op shader is always the
same.
This is required as we plan to add a default pipeline cache, that
would include our private shaders too.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 21 Jul 2020 22:51:19 +0000 (00:51 +0200)]
v3dv/pipeline_cache: MergePipelineCaches implementation
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 21 Jul 2020 21:26:26 +0000 (23:26 +0200)]
v3dv/pipeline_cache: support to serialize/deserialize cached NIRs
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Thu, 16 Jul 2020 10:41:11 +0000 (12:41 +0200)]
v3dv/pipeline_cache: cache v3dv_shader_variants
This also includes being able to serialize them as part of
GetPipelineCacheData and to deserialize it as part of
CreatePipelineCache.
So now we can also upload the assembly of the variant as part of the
PipelineCache creation.
Note that from all this the tricky part was the prog_data
serialization. v3d_compile allocates and fill a new prog_data, with
rzalloc. Among other things because it also allocates internally the
uniform list. So we needed to replicate that when deserializating the
prog_data. Ideally we would like to avoid that, and allocate as much
resources as possible using vk_alloc, but that would mean a somewhat
deep change on the v3d_compiler, that we want to avoid as much
possible for now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Sat, 18 Jul 2020 00:40:00 +0000 (02:40 +0200)]
v3dv/pipeline: add basic ref counting support for variants
As soon as we start to cache variants on pipeline caches, the same
variant could be used by different pipelines and pipeline caches.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 7 Jul 2020 13:56:44 +0000 (15:56 +0200)]
v3dv/pipeline_cache: cache nir shaders
Heavily based on anv nir caching. One of the bigger difference is that
we don't create the nir shader using a ralloc_context local to the
main compile graphics method. On anv, after compiling the shader, they
discard the nir shader. We need it as we could need it to build shader
variants later.
As anv, we introduce a environment variable to disable the cache:
V3DV_ENABLE_PIPELINE_CACHE
By default is enabled. The main purpose for this envvar is debugging,
in order to provide a easy way to discard a bug on the cache.
It is pending to serialize/deserialize the NIR shaders as part of
GetPipelineCacheData and PipelineCacheCreate. We also plan is to cache
too shader variants. We would do that on following patches.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Sat, 4 Jul 2020 11:10:49 +0000 (13:10 +0200)]
v3dv/pipeline_cache: bare basic support for pipeline cache
And this means providing a proper cache object, and being able to
load/retrieve a cache data with a proper header. Not really caching
anything yet. That would be tackle on following patches.
Note that this no-op cache got all the specific pipeline_cache and
pipeline.cache tests passing on the rpi4.
The following tests are still crashing when using the simulator:
dEQP-VK.synchronization.internally_synchronized_objects.pipeline_cache_compute
dEQP-VK.synchronization.internally_synchronized_objects.pipeline_cache_graphics
But those are an issue of synchronization tests on the simulator, and
not related with the pipeline cache itself. In general synchronization
tests should be tested on the rpi4.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Sat, 4 Jul 2020 11:09:58 +0000 (13:09 +0200)]
v3dv/device: proper pipeline cache uuid
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 7 Jul 2020 10:23:12 +0000 (12:23 +0200)]
v3dv/device: get proper device ID under simulator
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Fri, 3 Jul 2020 23:32:11 +0000 (01:32 +0200)]
v3dv/device: add vendorID/deviceID get helpers
As we would need them for the pipeline cache header.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 23 Jul 2020 10:45:41 +0000 (12:45 +0200)]
v3dv: fix dynamic state after meta operation
We should always save state on a push before starting a meta operation,
even if we don't have a pipeline, since dynamic state can be set at any
time directly on the command buffer. Similarly, we should always restore
it if the pop after the meta operation signals that it has written any
state, not only if we have a graphics pipeline to restore.
Fixes a rendering artifact in VkQuake.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 22 Jul 2020 08:12:51 +0000 (10:12 +0200)]
v3dv: enable fillModeNonSolid
This, together with the shaderClipDistance changes, gets vkQuake3
to run without hacks.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 21 Jul 2020 08:29:21 +0000 (10:29 +0200)]
v3dv: enable shaderClipDistance
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 16 Jul 2020 10:34:30 +0000 (12:34 +0200)]
v3dv: add a fast path for vkCmdClearAttachments
Since vkCmdClearAttachments executes inside a render pass, we would
benefit from converting it to a draw within the current subpass job to
improve batching and avoid expensive tile load/store operations.
This can dramatically improve performance for applications using this
command, however, we can only use this if we are clearing the base
layers of framebuffer attachments, since otherwise we would need to
use layered rendering, which we don't support yet.
This improves vkQuake3 performance dramatically (almost 100%
performance improvement at 1080p), which calls this twice per frame.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 17 Jul 2020 08:30:35 +0000 (10:30 +0200)]
v3dv: only clear depth/stencil attachments if any aspect needs clearing
When we can't perform the clears using the TLB and fallback to
vkCmdClearAttachments make sure we only emit the clear if any of
the depth/stencil aspects actually needs it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 17 Jul 2020 10:28:46 +0000 (12:28 +0200)]
v3dv: emit new shader state if viewport is dirty
If we have a dirty viewport then we need to re-emit our uniforms
to update QUNIFORM_VIEWPORT_*.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 16 Jul 2020 07:25:18 +0000 (09:25 +0200)]
v3dv: only care about barriers between GPU jobs
We don't need to do anything special for host/gpu synchronization.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 15 Jul 2020 07:31:39 +0000 (09:31 +0200)]
v3dv: avoid prime blit path when presenting WSI images
This allows us to present WSI images directly rather than going through
a blit path, improving performance.
This improves vkQuake performance at 1080p by 10-20%.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 14 Jul 2020 09:29:12 +0000 (11:29 +0200)]
v3dv: only use per-buffer clear bit for cases were we are already storing
This bit is helpful if we already need to store the buffer, but otherwise
we should not emit the store only to get the clear, we can use the global
clear packet for that and save us an expensive tile buffer store operation.
Also, we have not been using the per-buffer clear bit for depth/stencil
stores since "v3dv: fix depth/stencil clears on hardware", so we should
never been considering clearing needs to flag stores.
This improves vkQuake performance somewhere between 5%-15% by allowing
us to skip the store of the depth attachment.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 10 Jul 2020 07:50:57 +0000 (09:50 +0200)]
v3dv: ignore stencil load operation if attachment format doesn't have stencil
This gets vkQuake to render correctly, which creates render passes with
stencil load operations even when the depth/stencil attachment format
doesn't have a stencil aspect. While this is a bit weird, it seems to
be allowed by the spec:
"If the format has depth and/or stencil components, loadOp and storeOp
apply only to the depth data, while stencilLoadOp and stencilStoreOp
define how the stencil data is handled."
In our case we were not ignoring it and this was causing that we emitted a
Z buffer load that seemed to clobber the Z clear, preventing all draw calls
from passing the depth test.
While we are at it, also change the depth/stencil store operation (which
was already handling this scenario correctly) to use the format of the
render pass attachment description rather than the underlying image
format.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 9 Jul 2020 06:47:59 +0000 (08:47 +0200)]
v3dv: use a binning sync for CL jobs waiting on a semaphore
If we have a semaphore wait the job cannot be started before the semaphore
has been signaled, so we need to wait before starting the binning stage.
Fixes CTS failures in:
dEQP-VK.synchronization.op.single_queue.binary_semaphore.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 8 Jul 2020 07:56:44 +0000 (09:56 +0200)]
v3dv: improve pipeline barrier handling
So far we have been getting away with finishing the current job in the
presence of a pipeline barrier and relying on the RCL serialization,
but of course this is not always enough.
This patch addresses synchronization across different GPU units
(i.e. draw indirect after compute), as well as cases where we need to
sync before binning.
Fixes CTS failures in:
dEQP-VK.synchronization.op.single_queue.barrier.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 7 Jul 2020 11:10:44 +0000 (13:10 +0200)]
v3dv: don't support sRGB buffer formats
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 6 Jul 2020 09:44:00 +0000 (11:44 +0200)]
v3dv: handle VK_FORMAT_B8G8R8A8_UNORM vertex attributes
This is the only mandatory vertex attribute format that requires a
swizzle.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 7 Jul 2020 09:10:57 +0000 (11:10 +0200)]
v3d/compiler: support swapping R/B channels in vertex attributes.
We will need this in Vulkan to support vertex format
VK_FORMAT_B8G8R8A8_UNORM. The hardware doesn't allow to swizzle
vertex attribute components, so we need to do it in the shader.
v2:
- Use nir_intrinsic_io_semantics() to retrieve the location instead
of looping through the shader input variables (Eric).
- Assert that we only have one component (Eric).
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com> (v1)
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 7 Jul 2020 06:22:31 +0000 (08:22 +0200)]
v3dv: make sure we emit vertex attributes in location order
The order in which we emit the attributes is relevant, since
GL_SHADER_STATE_ATTRIBUTE_RECORD packets don't include an explicit
attribute index. This means that we need to emit them in driver
location order, since the compiler uses that location to compute
attribute offsets in the VPM.
Fixes ~1300 CTS tests in:
dEQP-VK.pipeline.vertex_input.multiple_attributes.out_of_order.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
n00b7 [Fri, 3 Jul 2020 08:46:51 +0000 (10:46 +0200)]
v3dv/device: handle primary nodes for newer kernels
For kernels rpi-5.X.y it changed from brcm,bcm2835-vc4 to
brcm,bcm2711-vc5, so let's just check for both strings.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 3 Jul 2020 08:06:28 +0000 (10:06 +0200)]
v3dv: fix color border clamping with specific formats
For some reason, CTS expects E5B9G9R9 and B10G11R11 with
transparent black border clamping produce alpha 1 instead of 0.
Since border color takes precedence over the texture state swizzle,
the only way to fix this is to lower the texture swizzle in the shader
to set alpha to 1.
Fixes:
dEQP-VK.pipeline.sampler.view_type.*b10g11r11*clamp_to_border_transparent_black
dEQP-VK.pipeline.sampler.view_type.*e5b9g9r9*.clamp_to_border_transparent_black
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 2 Jul 2020 10:43:42 +0000 (12:43 +0200)]
v3dv: fix regressions for cubemap array load/store
It seems that we only want to set the texture state's depth to the
number of 2D layers divided by 6 when sampling, not wen doing
load/store.
This means that we need to generate two different states and choose
the one to use depending on the descriptor.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 2 Jul 2020 08:56:05 +0000 (10:56 +0200)]
v3dv: use swizzle X001 with D/S formats
This matches Vulkan CTS expectations.
Fixes (for D/S formats):
dEQP-VK.pipeline.sampler.view_type.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Wed, 1 Jul 2020 15:03:28 +0000 (17:03 +0200)]
v3dv/image: fix TEXTURE_SHADER_STATE depth for cube arrays
For these we want to divide the number of layers by 6.
Fixes:
dEQP-VK.pipeline.image_view.view_type.cube_array.*
dEQP-VK.pipeline.image.suballocation.sampling_type.combined.view_type.cube_array.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 1 Jul 2020 12:21:09 +0000 (14:21 +0200)]
v3dv: handle unnormalized coordinates in samplers
In OpenGL, unnormalized coordinates are implicit based on the sampler
type (rectangle textures), so the compiler can set the flag when needed.
In Vulkan, however, this is configured explicitly in the sampler object,
so the compiler won't set it and we need to do it manually when we are
writing the P1 uniform.
Fixes:
dEQP-VK.pipeline.sampler.exact_sampling.*.unnormalized_coords
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 1 Jul 2020 09:05:28 +0000 (11:05 +0200)]
v3dv: fix blit_shader() to honor the region's aspect mask
When translating combined depth/stencil blits to compatible color blits we
should look at the requested region aspects to decide the color
mask to apply.
Fixes:
dEQP-VK.api.copy_and_blit.*.buffer_to_depthstencil.buffer_offset_d24_unorm_s8_uint_D
dEQP-VK.api.copy_and_blit.*.buffer_to_depthstencil.buffer_offset_d24_unorm_s8_uint_SD
dEQP-VK.api.copy_and_blit.*.buffer_to_depthstencil.buffer_offset_d24_unorm_s8_uint_S_D
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 30 Jun 2020 11:19:01 +0000 (13:19 +0200)]
v3dv: fix srcSubresource description for image to buffer blits
We are emitting a loop copying layer by layer here.
Fixes:
dEQP-VK.pipeline.render_to_image.core.*.huge.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 30 Jun 2020 08:44:43 +0000 (10:44 +0200)]
v3dv: fix textureSize() for cube arrays
For these we want to divide the number of layers by 6.
Fixes:
dEQP-VK.glsl.texture_functions.query.texturesize.*samplercubearray*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 30 Jun 2020 08:07:12 +0000 (10:07 +0200)]
v3dv: fix vkCmdCopyBuffer unaligned TLB access
We need to make sure that our tile load/store operations are aligned
to the selected item size for the copy.
Fixes simulator assert in VkQuake.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 29 Jun 2020 10:39:53 +0000 (12:39 +0200)]
v3dv: handle draw after barrier
Currently, we end the current job whenever the user emits a
pipeline barrier, but we then expect to have a valid job when
we emit a draw call.
If by the time we have to emit a draw call we don't have a valid
job, we need to create one by resuming execution of the current
subpass.
Fixes some tests in:
dEQP-VK.renderpass.suballocation.attachment_allocation.input_output.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 29 Jun 2020 09:56:48 +0000 (11:56 +0200)]
v3dv: improve handling of too large image sizes
Instead of asserting that users don't try to create images that
would require 4GB+ of memory, error out with the corresponding
OOM error when the user tries to actually allocate the memory
for the image.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 29 Jun 2020 09:17:19 +0000 (11:17 +0200)]
v3dv: add workaround for GFXH-1602
The first attribute must be active if using builtins.
This fixes a lot of simulator crashes for vertex input CTS tests.
It should be noted that some of these tests still fail after this
fix though, so there may be some other bug.
Fixes crashes in:
dEQP-VK.pipeline.vertex_input.multiple_attributes.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 29 Jun 2020 09:15:25 +0000 (11:15 +0200)]
v3dv: fix GFXH-930 workaround
We want to check if the coordinate shader is reading attributes, not
the vertex shader.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Sun, 28 Jun 2020 00:08:16 +0000 (02:08 +0200)]
v3dv/device: expose support for image cube array
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Mon, 29 Jun 2020 09:17:32 +0000 (11:17 +0200)]
nir: include texture query lod as one of the ops that requires a sampler
In practice we found that we need this for v3d (specifically for cube
map arrays, as they don't support the default value for wrap_i, so a
sampler object is needed to override that value).
It is worth to note that the main reason behind this auxiliar method
was to identify those cases that we didn't have a sampler object
available for Vulkan. So far, we found that we have a sampler object
coming from nir always for that operation.
Fixes cube map array tests like the following:
dEQP-VK.glsl.texture_functions.query.texturequerylod.usamplercubearray_fragment
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Wed, 24 Jun 2020 09:36:58 +0000 (11:36 +0200)]
v3dv/uniforms: support for some QUNIFORM_IMAGE_XXX
As we are using the descriptor sets to access to the specific
v3dv_image_view, we can just reuse the existing methods for the
textures.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 23 Jun 2020 23:55:20 +0000 (01:55 +0200)]
v3dv/pipeline: lower_image_deref
This is really similar to the existing lower_tex_src_to_offset, but
for now we prefer to keep them independent, just in case we start to
found specific image use-cases as we advance fixing CTS tests.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 23 Jun 2020 14:37:02 +0000 (16:37 +0200)]
v3dv/descriptors: support for DESCRIPTOR_TYPE_STORAGE_IMAGE
From the descriptor pov is mostly another image without sampler, so
for example, on the descriptor pool bo we only store the texture
shader state.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 26 Jun 2020 10:45:09 +0000 (12:45 +0200)]
v3dv: implement support for shader spilling
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 25 Jun 2020 08:49:13 +0000 (10:49 +0200)]
v3dv: lower frexp
Fixes:
dEQP-VK.glsl.builtin.precision.frexp.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 25 Jun 2020 08:28:07 +0000 (10:28 +0200)]
v3dv: lower unpack_{u,s}norm_2x16
The hardware doesn't have unorm/snorm packing variants and we were
already lowering the packing versions of these.
Fixes:
dEQP-VK.glsl.builtin.function.pack_unpack.unpacksnorm2x16_compute
dEQP-VK.glsl.builtin.function.pack_unpack.unpackunorm2x16_compute
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 25 Jun 2020 07:30:04 +0000 (09:30 +0200)]
v3dv: don't reset descriptor state after a meta operation
If the meta operation did not change descriptor state then we should keep it,
not reset it.
Fixes:
dEQP-VK.fragment_operations.early_fragment.early_fragment_tests_stencil
dEQP-VK.fragment_operations.early_fragment.no_early_fragment_tests_stencil
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 25 Jun 2020 06:54:34 +0000 (08:54 +0200)]
v3dv: handle empty set layouts
Fixes:
dEQP-VK.api.object_management.max_concurrent.descriptor_set_layout_empty
dEQP-VK.api.object_management.single_alloc_callbacks.descriptor_set_layout_empty
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 23 Jun 2020 13:47:53 +0000 (15:47 +0200)]
v3dv: initial support for input attachments
We are treating them as a special case of texture, so the commit is
mostly about integrating them with the existing
SAMPLER/SAMPLER_IMAGE/COMBINED_IMAGE_SAMPLER infrastructure.
This commit doesn't use in any special way the render pass
information, including the dependencies, so it is possible that we
would need to do something else. But this commit gets several CTS
tests, and two Sascha Willem Vulkan demos, so let's start with this
commit and handle any other use case for following commits.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Fri, 8 May 2020 23:06:12 +0000 (01:06 +0200)]
v3dv/pipeline: set load_layer_id to zero
We don't support layered rendering so let's lower the intrinsic to
zero for now. Based on turnip commit 99a4f7c ("turnip: set
load_layer_id to zero")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Mon, 22 Jun 2020 15:20:24 +0000 (17:20 +0200)]
v3dv/pipeline: iterate used textures using the combined index map
So far we were using nir->data.num_textures to fill the default values
for the textures used on the shader, and set the value for the number
of textures used.
But nir->data.num_textures doesn't take into account input
attachments, even after nir_lower_input_attachments. Although that
could make sense from a general pov, in our case we are treating input
attachments mostly as textures.
This commit count the number of textures interating through the
pipeline combined index map, as it includes both. This also makes the
populate of the shader key for default values more similar to the one
done at cmd_buffer with real values.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Mon, 22 Jun 2020 23:52:48 +0000 (01:52 +0200)]
v3d/tex: avoid to ask back for a sampler state if not needed
So far we were not asking the driver for the sampler state if we could
just use the default P1 values. But even if we need to fill P1 (for
example to fill up the output type of the format), if the texture
operation doesn't need a sampler, we can let that field as NULL (so
default values) and avoid calling back the driver for a sampler.
This is not mandatory for OpenGL (as we always have a sampler object),
although still a good to have. For Vulkan this is needed, as we don't
have a sampler object in that case.
v2: reword comment (Eric)
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 23 Jun 2020 10:46:49 +0000 (12:46 +0200)]
v3dv: pipeline initialization fixes for disabled rasterization
When rasterization is disabled there are a number of CreateInfo
structs that should be ignored. We were managing this correctly
for some cases, but not all of them. Specifically, viewport state
must be ignored and we weren't doing that.
Fixes:
dEQP-VK.api.descriptor_set.descriptor_set_layout_lifetime.graphics
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 23 Jun 2020 11:36:15 +0000 (13:36 +0200)]
v3dv: don't free BOs from imported memory objects
Only free the underlying BO when the exported memory object is freed
to avoid multiple frees of the same memory.
The only exception is winsys BOs where we import a BO created in the
display device into the render device. In this case, we only have one
memory object referencing the BO and we want to destroy it with that
memory object.
Fixes:
dEQP-VK.api.external.memory.dma_buf.*
dEQP-VK.api.external.memory.opaque_fd.*
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 23 Jun 2020 10:11:30 +0000 (12:11 +0200)]
v3dv: assert that our framebuffers are single sampled
This makes it easier to track CTS test fails related to multisampling,
which we don't support yet.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 23 Jun 2020 09:32:04 +0000 (11:32 +0200)]
v3dv: return a proper error for too large buffer allocations
Fixes: dEQP-VK.api.buffer.basic.size_max_uint64
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 19 Jun 2020 09:56:20 +0000 (11:56 +0200)]
v3dv: implement indirect compute dispatch
The hardware can't do this, so we need to record a CPU job that will
map the indirect buffer at queue submission time, read the dispatch
parameters and then submit a regular dispatch.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 19 Jun 2020 11:42:21 +0000 (13:42 +0200)]
v3dv: always emit index buffer state for new jobs
New jobs need to re-emit all state. Typically, this is achieved
by resetting all dirty state flags when we start a new job, but
for index buffers we were not using a dirty bit because we always
emit them immediately. This patch adds the bit and only tries
to skip index buffer state if the bit is not dirty, which will
ensure that we will always emit it for new jobs.
This fixes a regression in the shadowmapping demo from Sascha Willems
introduced with "v3dv: try harder to skip emission of redundant state".
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Fri, 19 Jun 2020 07:52:53 +0000 (09:52 +0200)]
v3dv: handle unsized arrays in SSBOs
CTS coverage for this was hiding behind compute shaders so
we didn't notice this was not working properly until now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 18 Jun 2020 11:53:51 +0000 (13:53 +0200)]
v3dv: implement compute dispatch
for now this only implements regular dispatches, not indirect.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 18 Jun 2020 10:14:58 +0000 (12:14 +0200)]
v3dv: handle separate binding points for compute and graphics
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 18 Jun 2020 09:06:00 +0000 (11:06 +0200)]
v3dv: support compute pipelines
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Thu, 18 Jun 2020 08:59:57 +0000 (10:59 +0200)]
v3dv/bo: add dump stats info
Focused specially on the cache, how many BOs and how much bo_size is
store on the cache, freed bos, bos moved to cache etc.
Initially not configured with V3D_DEBUG (like v3d) to avoid a runtime
check on most of v3dv_bo functions.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Tue, 16 Jun 2020 23:41:13 +0000 (01:41 +0200)]
v3dv/bo: add a maximum size for the bo_cache and a envvar to configure it
V3DV_MAX_BO_CACHE_SIZE can be used to configure it.
So one way to disable the bo cache is setting V3DV_MAX_BO_CACHE_SIZE
to zero. This would still run all the bo_cache size checks, but having
another envvar just to ensure that anything related to the cache is
used seemed like an overkill.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Alejandro Piñeiro [Fri, 5 Jun 2020 10:21:54 +0000 (12:21 +0200)]
v3dv/bo: adding a BO cache
Heavily based on the already existing for the v3d OpenGL driver, but
without references, and with some extra OOM checks (Vulkan CTS has
several OOM tests).
With this commit v3dv_bo_alloc and v3dv_bo_free became frontends to
the bo_cache. The former tries to get a BO from the cache if possible,
and the latter stores the BO on the cache if possible. The former also
adds a new parameter to point if the BO to allocate is private.
As v3d we are only caching private BOs, those created by the driver
for internal use (like CLs, tile_alloc, etc). They are the ones with
the highest change of being reused (for example, CL BOs are always
4KB, so they can always be reused). User-created BOs can have any
size, including some very large ones for buffers and images, which
makes them far less likely to be reused and would add a lot of memory
pressure if we decided to cache them.
In any case, in practice, we found that we could get a performance
improvement by caching also user-created BOs, but that would need more
care and an analysis to decide which ones makes sense. Would also
require to change how the cached BOs are stored by size. Right now
there are an array of list_head, that doesn't work well with big
BOs. If done, that would be handled on a separate commit.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 17 Jun 2020 10:15:42 +0000 (12:15 +0200)]
v3dv: always map full BOs
Both the API user and the driver may attempt to map a BO, possibly
only partially and using different ranges. This is a problem because
we only have a single map per BO. Fix this by making sure that when
a BO is mapped, we always map its entire range. This way if a BO
has been mapped before, we know that map is still valid no matter the
region we need to access now.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 16 Jun 2020 07:47:23 +0000 (09:47 +0200)]
v3dv: try to use TFU path when creating tiled images from linear buffers
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 15 Jun 2020 14:44:49 +0000 (16:44 +0200)]
v3dv: add a CPU path for buffer to image copies
The blit shader path for buffer to image copies is pretty bad,
since it needs to produce a tiled image from the linear buffer
prior to emitting the blit copy.
This patch adds a new preferential path where we implement the
copy using the CPU, similar to what the GL driver does for
texture uploads. This makes vkQuake2 at least 4x faster when
dynamic lights are enabled (which triggers dynamic texture
updates).
We also tested a GPU path where we use a shader that takes the
linear buffer as a UBO and copies directly from it. This also
shows a clear performance gain, but still worse than the CPU
implementation.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Mon, 15 Jun 2020 08:49:11 +0000 (10:49 +0200)]
v3dv: add a TFU path for buffer to image copies
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 11 Jun 2020 09:59:24 +0000 (11:59 +0200)]
v3dv: try harder to skip emission of redundant state
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Thu, 11 Jun 2020 06:32:26 +0000 (08:32 +0200)]
v3dv: actually enable early Z
We had done all the plumbing for this but EZ can be disabled in 3 places
and we were never setting the enable bit in the configuration bits packet.
Also, configuration bits must not enable EZ if this has been disabled in
the RCL for the whole frame, which we do if we don't have a depth
attachment at all.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Wed, 10 Jun 2020 10:12:21 +0000 (12:12 +0200)]
v3dv: fix release build warnings
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 9 Jun 2020 10:06:50 +0000 (12:06 +0200)]
v3dv: fix a few cases where we were ignoring suballocated buffers
This gets VkQuake2 to render correctly.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>
Iago Toral Quiroga [Tue, 9 Jun 2020 12:31:40 +0000 (14:31 +0200)]
v3dv: regen BO lists for CLs inside cloned jobs
Otherwise cloned BO lists point to the original list objects and not
the cloned ones, and that will confuse anything that tries to iterate over
them, such as list_length(), leading to infinite loops.
Fixes (in debug mode):
dEQP-VK.api.command_buffers.render_pass_continue
In that test we clone a full CL job from a secondary, and without this,
the BO lists in its CL lists will point to the bo_list field in the
original job, leading to an infinite loop as we assert the expected size
of these lists at queue submit time in handle_cl_job.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6766>