These implementations remove dependency on the Geometry pass by instead invoking one vertex shader instance for each expected output vertex, matching what a geometry shader would emit. Each vertex shader instance is then responsible for calculating the same output position based on its vertex_id as the logic would in the geometry shader version.
SSBO Vertex fetch enables full random-access into a vertex buffer by binding it as a read-only SSBO. This enables each instance to read neighbouring vertex data to perform contextual calculations as a geometry shader would, for cases where attribute Multiload is not supported.
Authored by Apple: Michael Parkin-White
Ref T96261
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D15901
MTLContext provides functionality for command encoding, binding management and graphics device management. MTLImmediate provides simple draw enablement with dynamically encoded data. These draws utilise temporary scratch buffer memory to provide minimal bandwidth overhead during workload submission.
This patch also contains empty placeholders for MTLBatch and MTLDrawList to enable testing of first pixels on-screen without failure.
The Metal API also requires access to the GHOST_Context to ensure the same pre-initialized Metal GPU device is used by the viewport. Given the explicit nature of Metal, explicit control is also needed over presentation, to ensure correct work scheduling and rendering pipeline state.
Authored by Apple: Michael Parkin-White
Ref T96261
(The diff is based on 043f59cb3b)
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D15953
Sun Disc is currently not supported because it'll need special handling - on the one hand, I'm not sure if Eevee would handle a 1e6 coming out of a background shader without issues, and on the other hand it won't actually cast sharp shadows anyways.
I guess we'd want to internally add a sun to the lamps if Sun Disc is enabled, but getting that right is tricky since the user could e.g. swap RGB channels in the node tree and the lamp wouldn't match that.
Anyways, that can be handled later, the sky itself is already a start.
Reviewed By: fclem
Differential Revision: https://developer.blender.org/D13522
Whether faces are hidden and face sets are orthogonal concepts, but
currently sculpt mode stores them together in the face set array.
This means that if anything is hidden, there must be face sets,
and if there are face sets, we have to keep track of what is hidden.
In other words, it adds a bunch of redundant work and state tracking.
On the user level it's nice that face sets and hiding are consistent,
but we don't need to store them together to accomplish that.
This commit uses the `".hide_poly"` attribute from rB2480b55f216c to
read and change hiding in sculpt mode. Face sets don't need to be
negative anymore, and a bunch of "face set <-> hide status" conversion
can be removed. Plus some other benefits:
- We don't need to allocate either array quite as much.
- The hide status can be read from 1/4 the memory as face sets.
- Updates when entering or exiting sculpt mode can be removed.
- More opportunities for early-outs when nothing is hidden.
- Separating concerns makes sculpt code more obvious.
- It will be easier to convert face sets into a generic int attribute.
Differential Revision: https://developer.blender.org/D15950
This change adds cryptomatte render passes to EEVEE-Next. Due to the upcoming viewport
compositor we also improved cryptomatte so it will be real-time. This also allows viewing
the cryptomatte passes in the viewport directly.
{F13482749}
A surface shader would store any active cryptomatte layer to a texture. Object hash is stored
as R, Asset hash as G and Material hash as B. Hashes are only calculated when the cryptomatte
layer is active to reduce any unneeded work.
During film accumulation the hashes are separated and stored in a texture array that matches
the cryptomatte standard. For the real-time use case sorting is skipped. For final rendering
the samples are sorted and normalized.
NOTE: Eventually we should also do sample normalization in the viewport in order to extract the correct
mask when using the viewport compositor.
Reviewed By: fclem
Maniphest Tasks: T99390
Differential Revision: https://developer.blender.org/D15753
corresponding data layers and using their values for computations.
Avoiding that should increase performance in many operations that
would otherwise have to read, write, or propagate these values.
It also means decreased memory usage-- not just for sculpt mode
but for any mesh that was in sculpt mode. Previously the mask, face set,
and hide status layers were *always* allocated by sculpt mode.
Here are a few basic tests when masking and face sets are not used:
| Test | Before | After |
| Subsurf Modifier | 148 ms | 126 ms |
| Sculpt Overlay Extraction | 24 ms every redraw | 0 ms |
| Memory usage | 252 MB | 236 MB |
I wouldn't expect any difference when they are used though.
The code changes are mostly just making sculpt features safe for when
the layers aren't stored, and some changes to the conversion to and
from the hide layers. Use of the ".hide_poly" attribute replaces testing
whether face sets are negative in many places.
Differential Revision: https://developer.blender.org/D15937
Use `verts` instead of `vertices` and `polys` instead of `polygons`
in the API added in 05952aa94d. This aligns better with
existing naming where the shorter names are much more common.
The only difference between `GPU_SHADER_2D_LINE_DASHED_UNIFORM_COLOR`
and `GPU_SHADER_3D_LINE_DASHED_UNIFORM_COLOR` is that in the vertex
shader the 2D version uses `vec4(pos, 0.0, 1.0)` and the 3D version
uses `vec4(pos, 1.0)`.
But VBOs with 2D attributes work perfectly in shaders that use 3D
attributes. Components not specified are filled with components from
`vec4(0.0, 0.0, 0.0, 1.0)`.
So there is no real benefit to having two different shader versions.
The only real difference between `GPU_SHADER_2D_SMOOTH_COLOR` and
`GPU_SHADER_3D_SMOOTH_COLOR` is that in the vertex shader the 2D
version uses `vec4(pos, 0.0, 1.0)` and the 3D version uses
`vec4(pos, 1.0)`.
But VBOs with 2D attributes work perfectly in shaders that use 3D
attributes. Components not specified are filled with components from
`vec4(0.0, 0.0, 0.0, 1.0)`.
So there is no real benefit to having two different shader versions.
This will simplify porting shaders to python as it will not be
necessary to use a 3D and a 2D version of the shaders.
In python the new name for '2D_SMOOTH_COLOR' and '3D_SMOOTH_COLOR'
is 'SMOOTH_COLOR', but the old names still work for backward
compatibility.
The only real difference between `GPU_SHADER_2D_IMAGE` and
`GPU_SHADER_3D_IMAGE` is that in the vertex shader the 2D
version uses `vec4(pos, 0.0, 1.0)` and the 3D version uses
`vec4(pos, 1.0)`.
But VBOs with 2D attributes work perfectly in shaders that use 3D
attributes. Components not specified are filled with components from
`vec4(0.0, 0.0, 0.0, 1.0)`.
So there is no real benefit to having two different shader versions.
This will simplify porting shaders to python as it will not be
necessary to use a 3D and a 2D version of the shaders.
In python the new name for '2D_IMAGE' and '3D_IMAGE'
is 'IMAGE', but the old names still work for backward
compatibility.
The only real difference between `GPU_SHADER_2D_FLAT_COLOR` and
`GPU_SHADER_3D_FLAT_COLOR` is that in the vertex shader the 2D
version uses `vec4(pos, 0.0, 1.0)` and the 3D version uses
`vec4(pos, 1.0)`.
But VBOs with 2D attributes work perfectly in shaders that use 3D
attributes. Components not specified are filled with components from
`vec4(0.0, 0.0, 0.0, 1.0)`.
So there is no real benefit to having two different shader versions.
This will simplify porting shaders to python as it will not be
necessary to use a 3D and a 2D version of the shaders.
In python the new name for '2D_FLAT_COLOR'' and '3D_FLAT_COLOR'
is 'FLAT_COLOR', but the old names still work for backward
compatibility.
The only real difference between `GPU_SHADER_2D_UNIFORM_COLOR` and
`GPU_SHADER_3D_UNIFORM_COLOR` is that in the vertex shader the 2D
version uses `vec4(pos, 0.0, 1.0)` and the 3D version uses
`vec4(pos, 1.0)`.
But VBOs with 2D attributes work perfectly in shaders that use 3D
attributes. Components not specified are filled with components from
`vec4(0.0, 0.0, 0.0, 1.0)`.
So there is no real benefit to having two different shader versions.
This will simplify porting shaders to python as it will not be
necessary to use a 3D and a 2D version of the shaders.
In python the new name for '2D_UNIFORM_COLOR'' and '3D_UNIFORM_COLOR'
is 'UNIFORM_COLOR', but the old names still work for backward
compatibility.
Differential Revision: https://developer.blender.org/D15836
For copy-on-write, we want to share attribute arrays between meshes
where possible. Mutable pointers like `Mesh.mvert` make that difficult
by making ownership vague. They also make code more complex by adding
redundancy.
The simplest solution is just removing them and retrieving layers from
`CustomData` as needed. Similar changes have already been applied to
curves and point clouds (e9f82d3dc7, 410a6efb74). Removing use of
the pointers generally makes code more obvious and more reusable.
Mesh data is now accessed with a C++ API (`Mesh::edges()` or
`Mesh::edges_for_write()`), and a C API (`BKE_mesh_edges(mesh)`).
The CoW changes this commit makes possible are described in T95845
and T95842, and started in D14139 and D14140. The change also simplifies
the ongoing mesh struct-of-array refactors from T95965.
**RNA/Python Access Performance**
Theoretically, accessing mesh elements with the RNA API may become
slower, since the layer needs to be found on every random access.
However, overhead is already high enough that this doesn't make a
noticible differenc, and performance is actually improved in some
cases. Random access can be up to 10% faster, but other situations
might be a bit slower. Generally using `foreach_get/set` are the best
way to improve performance. See the differential revision for more
discussion about Python performance.
Cycles has been updated to use raw pointers and the internal Blender
mesh types, mostly because there is no sense in having this overhead
when it's already compiled with Blender. In my tests this roughly
halves the Cycles mesh creation time (0.19s to 0.10s for a 1 million
face grid).
Differential Revision: https://developer.blender.org/D15488
This replaces the direct shader uniform layout declaration by a linear
search through a global buffer.
Each instance has an attribute offset inside the global buffer and an
attribute count.
This removes any padding and tighly pack all uniform attributes inside
a single buffer.
This would also remove the limit of 8 attribute but it is kept because of
compatibility with the old system that is still used by the old draw
manager.
This is a new implementation of the draw manager using modern
rendering practices and GPU driven culling.
This only ports features that are not considered deprecated or to be
removed.
The old DRW API is kept working along side this new one, and does not
interfeer with it. However this needed some more hacking inside the
draw_view_lib.glsl. At least the create info are well separated.
The reviewer might start by looking at `draw_pass_test.cc` to see the
API in usage.
Important files are `draw_pass.hh`, `draw_command.hh`,
`draw_command_shared.hh`.
In a nutshell (for a developper used to old DRW API):
- `DRWShadingGroups` are replaced by `Pass<T>::Sub`.
- Contrary to DRWShadingGroups, all commands recorded inside a pass or
sub-pass (even binds / push_constant / uniforms) will be executed in order.
- All memory is managed per object (except for Sub-Pass which are managed
by their parent pass) and not from draw manager pools. So passes "can"
potentially be recorded once and submitted multiple time (but this is
not really encouraged for now). The only implicit link is between resource
lifetime and `ResourceHandles`
- Sub passes can be any level deep.
- IMPORTANT: All state propagate from sub pass to subpass. There is no
state stack concept anymore. Ensure the correct render state is set before
drawing anything using `Pass::state_set()`.
- The drawcalls now needs a `ResourceHandle` instead of an `Object *`.
This is to remove any implicit dependency between `Pass` and `Manager`.
This was a huge problem in old implementation since the manager did not
know what to pull from the object. Now it is explicitly requested by the
engine.
- The pases need to be submitted to a `draw::Manager` instance which can
be retrieved using `DRW_manager_get()` (for now).
Internally:
- All object data are stored in contiguous storage buffers. Removing a lot
of complexity in the pass submission.
- Draw calls are sorted and visibility tested on GPU. Making more modern
culling and better instancing usage possible in the future.
- Unit Tests have been added for regression testing and avoid most API
breakage.
- `draw::View` now contains culling data for all objects in the scene
allowing caching for multiple views.
- Bounding box and sphere final setup is moved to GPU.
- Some global resources locations have been hardcoded to reduce complexity.
What is missing:
- ~~Workaround for lack of gl_BaseInstanceARB.~~ Done
- ~~Object Uniform Attributes.~~ Done (Not in this patch)
- Workaround for hardware supporting a maximum of 8 SSBO.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D15817
This particular GPU driver does not constant fold all the way in order
to discard the unused branches.
To workaround that, we introduce a series of material flag that generates
defines that only keep used branches.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D15852
Full support for translation and compilation of shaders in Metal, using
GPUShaderCreateInfo. Includes render pipeline state creation and management,
enabling all standard GPU viewport rendering features in Metal.
Authored by Apple: Michael Parkin-White, Marco Giordano
Ref T96261
Reviewed By: fclem
Maniphest Tasks: T96261
Differential Revision: https://developer.blender.org/D15563
- Adding in compatibility paths to support minimum per-vertex strides for vertex formats. OpenGL supports a minimum stride of 1 byte, in Metal, this minimum stride is 4 bytes. Meaing a vertex format must be atleast 4-bytes in size.
- Replacing transform feedback compile-time check to conditional look-up, given TF is supported on macOS with Metal.
- 3D texture size safety check added as a general capability, rather than being in the gl backend only. Also required for Metal.
Authored by Apple: Michael Parkin-White
Ref T96261
Reviewed By: fclem
Maniphest Tasks: T96261
Differential Revision: https://developer.blender.org/D14510
Implementation also contains a number of optimisations and feature enablements specific to the Metal API and Apple Silicon GPUs.
Ref T96261
Reviewed By: fclem
Maniphest Tasks: T96261
Differential Revision: https://developer.blender.org/D15369
With the new `attrs_info_get` method, we can get information about
the attributes used in a `GPUShader` and thus have more freedom in the
automatic creation of `GPUVertFormat`s
Reviewed By: fclem, campbellbarton
Differential Revision: https://developer.blender.org/D15764
Push the const usage a bit further for compositor nodes, so that they
are more explicit about not modifying original nodes from the editor.
Differential Revision: https://developer.blender.org/D15822
This patch moves material indices from the mesh `MPoly` struct to a
generic integer attribute. The builtin material index was already
exposed in geometry nodes, but this makes it a "proper" attribute
accessible with Python and visible in the "Attributes" panel.
The goals of the refactor are code simplification and memory and
performance improvements, mainly because the attribute doesn't have
to be stored and processed if there are no materials. However, until
4.0, material indices will still be read and written in the old
format, meaning there may be a temporary increase in memory usage.
Further notes:
* Completely removing the `MPoly.mat_nr` after 4.0 may require
changes to DNA or introducing a new `MPoly` type.
* Geometry nodes regression tests didn't look at material indices,
so the change reveals a bug in the realize instances node that I fixed.
* Access to material indices from the RNA `MeshPolygon` type is slower
with this patch. The `material_index` attribute can be used instead.
* Cycles is changed to read from the attribute instead.
* BMesh isn't changed in this patch. Theoretically it could be though,
to save 2 bytes per face when less than two materials are used.
* Eventually we could use a 16 bit integer attribute type instead.
Ref T95967
Differential Revision: https://developer.blender.org/D15675
This checks for the availability of `gl_BaseInstanceARB` or equivalent.
Disabling for any workaround that disables shader_image_load_store_support
as a preventive measure.
This allows the render engine to expect non-overlapping resources in the
generated create info.
Textures are indexed from 0 and up.
Nodetree ubo is bound to slot 0.
Uniform attributes ubo is bound to slot 1.
The number of attribute domains isn't an attribute domain, so storing
ATTR_DOMAIN_NUM in a variable with an eAttrDomain type isn't correct.
In the cases it was used, the value wouldn't be accessed anyway.
There was a memory leak in the GPU code generator for the compositor
output. It was just due to a missing free in the GPU code generator
destructor, so this patch makes sure it is freed.
With libepoxy we can choose between EGL and GLX at runtime, as well as
dynamically open EGL and GLX libraries without linking to them.
This will make it possible to build with Wayland, EGL, GLVND support while
still running on systems that only have X11, GLX and libGL. It also paves
the way for headless rendering through EGL.
libepoxy is a new library dependency, and is included in the precompiled
libraries. GLEW is no longer a dependency, and WITH_SYSTEM_GLEW was removed.
Includes contributions by Brecht Van Lommel, Ray Molenkamp, Campbell Barton
and Sergey Sharybin.
Ref T76428
Differential Revision: https://developer.blender.org/D15291
