Age | Commit message (Collapse) | Author |
|
This patch adds the Metal host-side code:
- Add all core host-side Metal backend files (device_impl, queue, etc)
- Add MetalRT BVH setup files
- Integrate with Cycles device enumeration code
- Revive `path_source_replace_includes` in util/path (required for MSL compilation)
This patch also includes a couple of small kernel-side fixes:
- Add an implementation of `lgammaf` for Metal [Nemes, Gergő (2010), "New asymptotic expansion for the Gamma function", Archiv der Mathematik](https://users.renyi.hu/~gergonemes/)
- include "work_stealing.h" inside the Metal context class because it accesses state now
Ref T92212
Reviewed By: brecht
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D13423
|
|
|
|
Found in D13353, likely has no significant impact in performance.
|
|
This patch adds MetalRT support to Cycles kernel code. It is mostly additive in nature or confined to Metal-specific code, however there are a few areas where this interacts with other code:
- MetalRT closely follows the Optix implementation, and in some cases (notably handling of transforms) it makes sense to extend Optix special-casing to MetalRT. For these generalisations we now have `__KERNEL_GPU_RAYTRACING__` instead of `__KERNEL_OPTIX__`.
- MetalRT doesn't support primitive offsetting (as with `primitiveIndexOffset` in Optix), so we define and populate a new kernel texture, `__object_prim_offset`, containing per-object primitive / curve-segment offsets. This is referenced and applied in MetalRT intersection handlers.
- Two new BVH layout enum values have been added: `BVH_LAYOUT_METAL` and `BVH_LAYOUT_MULTI_METAL_EMBREE` for XPU mode). Some host-side enum case handling has been updated where it is trivial to do so.
Ref T92212
Reviewed By: brecht
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D13353
|
|
This patch fixes an address space mismatch in the film convert kernels on Metal. The `film_get_pass_pixel_...` functions take a `ccl_private` result pointer, but the film convert kernels pass a `ccl_global` memory pointer. Specialising the pass-fetch functions with templates results in compilation errors on Visual Studio, so instead this patch just adds an intermediate local on Metal.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D13350
|
|
Fix performance decrease with Scrambling Distance on
|
|
With the current code in master, scrambling distance is enabled on non-hardware accelerated ray tracing devices see a measurable performance decrease when compared scrambling distance on vs off. From testing, this performance decrease comes from the large tile sizes scheduled in `tile.cpp`.
This patch attempts to address the performance decrease by using different algorithms to calculate the tile size for devices with hardware accelerated ray traversal and devices without. Large tile sizes for hardware accelerated devices and small tile sizes for others.
Most of this code is based on proposals from @brecht and @leesonw
Reviewed By: brecht, leesonw
Differential Revision: https://developer.blender.org/D13042
|
|
catmull-rom curve type
Some enum names were changed/removed in OptiX 7.4, so some changes are necessary to
make things compile still.
In addition, OptiX 7.4 also adds built-in support for catmull-rom curves, so it is no longer
necessary to convert the catmull-rom data to cubic bsplines first, and has endcaps disabled
by default now, so can remove the special handling via any-hit programs that filtered them
out before.
Differential Revision: https://developer.blender.org/D13351
|
|
Contributed by luzpaz.
Differential Revision: https://developer.blender.org/D10447
|
|
MSL requires that constant address space literals be declared at program
scope. This patch moves the `blackbody_table_r/g/b` and `cie_colour_match`
constants into separate files so they can be declared at the appropriate scope.
Ref T92212
Differential Revision: https://developer.blender.org/D13241
|
|
This patch contains many small leftover fixes and additions that are
required for Metal-enablement:
- Address space fixes and a few other small compile fixes
- Addition of missing functionality to the Metal adapter headers
- Addition of various scattered `__KERNEL_METAL__` blocks (e.g. for
atomic support & maths functions)
Ref T92212
Differential Revision: https://developer.blender.org/D13263
|
|
Depends on D13243
Differential Revision: https://developer.blender.org/D13244
|
|
Introduce a packed_float3 type for smaller storage that is exactly 3
floats, instead of 4. For computation float3 is still used since it can
use SIMD instructions.
Ref T92212
Differential Revision: https://developer.blender.org/D13243
|
|
This patch adapts the existing volumetric read/write lambda functions for Metal. Lambda expressions are not supported on MSL, so two new macros `VOLUME_READ_LAMBDA` and `VOLUME_WRITE_LAMBDA` have been defined with a default implementation which, on Metal, is overridden to use inline function objects.
This patch also removes the last remaining mention of the now-unused `ccl_addr_space`.
Ref T92212
Reviewed By: leesonw
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D13234
|
|
|
|
|
|
|
|
The issue was caused by splitting happening twice.
Fixed by checking for split flag which is assigned to the both states
during split.
The tricky part was to write catcher data at the moment of split: the
transparency and shadow catcher sample count is to be accumulated at
that point. Now it is happening in the `intersect_closest` kernel.
The downside is that render buffer is to be passed to the kernel, but
the benefit is that extra split bounce check is not needed now.
Had to move the passes write to shadow catcher header, since include
of `film/passes.h` causes all the fun of requirement to have BSDF
data structures available.
Differential Revision: https://developer.blender.org/D13177
|
|
|
|
It's unclear why this fails. Maybe the size of half4 is not the expected
8 bytes and adjacent pixels are overwritten. Or there is some bug in the
HIP compiler writing a struct into global memory, which we probably don't
do elsewhere in the kernel.
Thanks to Thomas, William and Jeroen for helping investigate this.
|
|
rB3a4c8f406a3a3bf0627477c6183a594fa707a6e2 changed the macros that create the film
convert kernel entry points, but in the process accidentally changed the parameter definition
to one of those (which caused CUDA launch and misaligned address errors) and changed the
implementation as well. This restores the correct implementation from before.
In addition, the `ccl_gpu_kernel_threads` macro did not work as intended and caused the
generated launch bounds to end up with an incorrect input for the second parameter (it was
set to "thread_num_registers", rather than the result of the block number calculation). I'm
not entirely sure why, as the macro definition looked sound to me. Decided to simply go with
two separate macros instead, to simplify and solve this.
Also changed how state is captured with the `ccl_gpu_kernel_lambda` macro slightly, to avoid
a compiler warning (expression has no effect) that otherwise occurred.
Maniphest Tasks: T92985
Differential Revision: https://developer.blender.org/D13175
|
|
This patch adapts the shared kernel entrypoints so that they can be compiled as MSL (Metal Shading Language). Where possible, the adaptations avoid changes in common code.
In MSL, kernel function inputs are explicitly bound to resources. In the case of argument buffers, we declare a struct containing the kernel arguments, accessible via device pointer. This differs from CUDA and HIP where kernel function arguments are declared as traditional C-style function parameters. This patch adapts the entrypoints declared in kernel.h so that they can be translated via a new `ccl_gpu_kernel_signature` macro into the required parameter struct + kernel entrypoint pairing for MSL.
MSL buffer attribution must be applied to function parameters or non-static class data members. To allow universal access to the integrator state, kernel data, and texture fetch adapters, we wrap all of the shared kernel code in a `MetalKernelContext` class. This is achieved by bracketing the appropriate kernel headers with "context_begin.h" and "context_end.h" on Metal. When calling deeper into the kernel code, we must reference the context class (e.g. `context.integrator_init_from_camera`). This extra prefixing is performed by a set of defines in "context_end.h". These will require explicit maintenance if entrypoints change. We invite discussion on more maintainable ways to enforce correctness.
Lambda expressions are not supported on MSL, so a new `ccl_gpu_kernel_lambda` macro generates an inline function object and optionally capturing any required state. This yields the same behaviour. This approach is applied to all parallel_... implementations which are templated by operation. The lambda expressions in the film_convert... kernels don't adapt cleanly to use function objects. However, these entrypoints can be macro-generated more concisely to avoid lambda expressions entirely, instead relying on constant folding to handle the pixel/channel conversions.
A separate implementation of `gpu_parallel_active_index_array` is provided for Metal to workaround some subtle differences in SIMD width, and also to encapsulate some required thread parameters which must be declared as explicit entrypoint function parameters.
Ref T92212
Reviewed By: brecht
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D13109
|
|
is destroyed
Adds a pass before denoising that calculates the intensity of the image, which can be
passed into the OptiX denoiser for more optimal results for very dark or very bright images.
In addition this also fixes a crash that sometimes occurred on exit. The OptiX denoiser object
has to be destroyed before the OptiX device context object (since it references that). But in
C++ the destructor function of a class is called before its fields are destructed, so
"~OptiXDevice" was always called before "OptiXDevice::~Denoiser" and therefore
"optixDeviceContextDestroy" was called before "optixDenoiserDestroy", hence the crash.
Differential Revision: https://developer.blender.org/D13160
|
|
Adds a bunch of CPU kernel function to process on row of pixels, and use those
instead of calling unoptimized implementations.
Fixes T92598
|
|
Remove prefix of filenames that is the same as the folder name. This used
to help when #includes were using individual files, but now they are always
relative to the cycles root directory and so the prefixes are redundant.
For patches and branches, git merge and rebase should be able to detect the
renames and move over code to the right file.
|
|
* Split render/ into scene/ and session/. The scene/ folder now contains the
scene and its nodes. The session/ folder contains the render session and
associated data structures like drivers and render buffers.
* Move top level kernel headers into new folders kernel/camera/, kernel/film/,
kernel/light/, kernel/sample/, kernel/util/
* Move integrator related kernel headers into kernel/integrator/
* Move OSL shaders from kernel/shaders/ to kernel/osl/shaders/
For patches and branches, git merge and rebase should be able to detect the
renames and move over code to the right file.
|
|
|
|
* Additional structs added to the hipew loader for device props
* Adds hipRTC functions to the loader for future usage
* Enables CPU+GPU usage for HIP
* Cleanup to the adaptive kernel compilation process
* Fix for kernel compilation failures with HIP with latest master
Ref T92393, D12958
|
|
Similar to main path compaction that happens before adding work tiles, this
compacts shadow paths before launching kernels that may add shadow paths.
Only do it when more than 50% of space is wasted.
It's not a clear win in all scenes, some are up to 1.5% slower. Likely caused
by different order of scheduling kernels having an unpredictable performance
impact. Still feels like compaction is just the right thing to avoid cases
where a few shadow paths can hold up a lot of main paths.
Differential Revision: https://developer.blender.org/D12944
|
|
|
|
Taking advantage of the new decoupled main and shadow paths. For CPU we
just store two nested structs in the integrator state, one for direct light
shadows and one for AO. For the GPU we restrict the number of shade surface
states to be executed based on available space in the shadow paths queue.
This also helps improve performance in benchmark scenes with an AO pass,
since it is no longer needed to use the shader raytracing kernel there,
which has worse performance.
Differential Revision: https://developer.blender.org/D12900
|
|
These transparent shadows can be expansive to evaluate. Especially on the
GPU they can lead to poor occupancy when only some pixels require many kernel
launches to trace and evaluate many layers of transparency.
Baked transparency allows tracing a single ray in many cases by accumulating
the throughput directly in the intersection program without recording hits
or evaluating shaders. Transparency is baked at curve vertices and
interpolated, for most shaders this will look practically the same as actual
shader evaluation.
Fixes T91428, performance regression with spring demo file due to transparent
hair, and makes it render significantly faster than Blender 2.93.
Differential Revision: https://developer.blender.org/D12880
|
|
Helps save one OptiX payload and is a bit more efficient.
Differential Revision: https://developer.blender.org/D12909
|
|
The motivation for this is twofold. It improves performance (5-10% on most
benchmark scenes), and will help to bring back transparency support for the
ambient occlusion pass.
* Duplicate some members from the main path state in the shadow path state.
* Add shadow paths incrementally to the array similar to what we do for
the shadow catchers.
* For the scheduling, allow running shade surface and shade volume kernels
as long as there is enough space in the shadow paths array. If not, execute
shadow kernels until it is empty.
* Add IntegratorShadowState and ConstIntegratorShadowState typedefs that
can be different between CPU and GPU. For GPU both main and shadow paths
juse have an integer for SoA access. Bt with CPU it's a different pointer
type so we get type safety checks in code shared between CPU and GPU.
* For CPU, add a separate IntegratorShadowStateCPU struct embedded in
IntegratorShadowState.
* Update various functions to take the shadow state, and make SVM take either
type of state using templates.
Differential Revision: https://developer.blender.org/D12889
|
|
Only copy the number of items used instead of the max items.
Ref D12889
|
|
* Rename struct KernelGlobals to struct KernelGlobalsCPU
* Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs
that every device can define in its own way.
* Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and
replace with these new typedefs.
* Add explicit state argument to INTEGRATOR_STATE and similar macros
In preparation for decoupling main and shadow paths.
Differential Revision: https://developer.blender.org/D12888
|
|
|
|
|
|
|
|
This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation.
MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness.
The vast majority of deltas in this patch fall into one of two cases:
- Ensuring ccl_private is specified for thread-local pointer types
- Ensuring ccl_global is specified for device-wide pointer types
Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant.
In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture.
The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation.
Ref T92212
Reviewed By: brecht
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D12864
|
|
Previously the storage here was optimized to avoid indirections in BVH2
traversal. This helps improve performance a bit, but makes performance
and memory usage of Embree and OptiX BVHs a bit worse also. It also adds
code complexity in other parts of the code.
Now decouple triangle and curve primitive storage from BVH2.
* Reduced peak memory usage on all devices
* Bit better performance for OptiX and Embree
* Bit worse performance for CUDA
* Simplified code:
** Intersection.prim/object now matches ShaderData.prim/object
** No more offset manipulation for mesh displacement before a BVH is built
** Remove primitive packing code and flags for Embree and OptiX
** Curve segments are now stored in a KernelCurve struct
* Also happens to fix a bug in baking with incorrect prim/object
Fixes T91968, T91770, T91902
Differential Revision: https://developer.blender.org/D12766
|
|
Implement an overscan support for tiles, so that adaptive sampling can
rely on the pixels neighbourhood.
Differential Revision: https://developer.blender.org/D12599
|
|
|
|
NOTE: this feature is not ready for user testing, and not yet enabled in daily
builds. It is being merged now for easier collaboration on development.
HIP is a heterogenous compute interface allowing C++ code to be executed on
GPUs similar to CUDA. It is intended to bring back AMD GPU rendering support
on Windows and Linux.
https://github.com/ROCm-Developer-Tools/HIP.
As of the time of writing, it should compile and run on Linux with existing
HIP compilers and driver runtimes. Publicly available compilers and drivers
for Windows will come later.
See task T91571 for more details on the current status and work remaining
to be done.
Credits:
Sayak Biswas (AMD)
Arya Rafii (AMD)
Brian Savery (AMD)
Differential Revision: https://developer.blender.org/D12578
|
|
Before the visibility test against the visibility flags was performed in an any-hit program in OptiX
(called `__anyhit__kernel_optix_visibility_test`), which was using the `__prim_visibility` array.
This is not entirely correct however, since `__prim_visibility` is filled with the merged visibility
flags of all objects that reference that primitive, so if one object uses different visibility flags
than another object, but they both are instances of the same geometry, they would appear the same
way. The reason that the any-hit program was used rather than the OptiX instance visibility mask is
that the latter is currently limited to 8 bits only, which is not sufficient to contain all Cycles
visibility flags (12 bits).
To mostly fix the problem with multiple instances and different visibility flags, I changed things to
use the OptiX instance visibility mask for a subset of the Cycles visibility flags (`PATH_RAY_CAMERA`
to `PATH_RAY_VOLUME_SCATTER`, which fit into 8 bits) and only fall back to the visibility test any-hit
program if that isn't enough (e.g. the ray visibility mask exceeds 8 bits or when using the built-in
curves from OptiX, since the any-hit program is then also used to skip the curve endcaps).
This may also improve performance in some cases, since by default OptiX can now perform the normal
scene intersection trace calls entirely on RT cores without having to jump back to the SM on every
hit to execute the any-hit program.
Fixes T89801
Differential Revision: https://developer.blender.org/D12604
|
|
|
|
|
|
This includes much improved GPU rendering performance, viewport interactivity,
new shadow catcher, revamped sampling settings, subsurface scattering anisotropy,
new GPU volume sampling, improved PMJ sampling pattern, and more.
Some features have also been removed or changed, breaking backwards compatibility.
Including the removal of the OpenCL backend, for which alternatives are under
development.
Release notes and code docs:
https://wiki.blender.org/wiki/Reference/Release_Notes/3.0/Cycles
https://wiki.blender.org/wiki/Source/Render/Cycles
Credits:
* Sergey Sharybin
* Brecht Van Lommel
* Patrick Mours (OptiX backend)
* Christophe Hery (subsurface scattering anisotropy)
* William Leeson (PMJ sampling pattern)
* Alaska (various fixes and tweaks)
* Thomas Dinges (various fixes)
For the full commit history, see the cycles-x branch. This squashes together
all the changes since intermediate changes would often fail building or tests.
Ref T87839, T87837, T87836
Fixes T90734, T89353, T80267, T80267, T77185, T69800
|