| Age | Commit message (Collapse) | Author |
|---|
|
The approach didn't turn out to be that useful after all as there
are unavoidable dependencies of data from the device. For example,
to know whether object intersects volume or not it is required to
run displacement kernels.
The way of splitting host and device updates caused state where
some data is not yet available, causing confusion and leaving
code to be error-prone.
|
|
|
|
|
|
|
|
* 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
|
|
Mention required CUDA and OptiX compute capability and minimum driver
version. For HIP there is a placeholder until we know the supported
architectures.
|
|
This triggered a compiler bug where it does not handle the sub.s16 PTX
instruction. Instead refactor the code so we don't need to do uint16_t
subtraction at all.
Also update OptiX device to remove the AO pass direct callable.
Thanks Patrick Mours for figuring this out.
|
|
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
|
|
Ref D12834
|
|
Apple's international keyboards have an additional `kVK_ISO_Section`
key. With some (Italian, Spanish) keyboard layouts, this is `\`, `[` keys
which Blender keymap can use.
Right now this key is explicitly set as `Unknown`.
Note that `kVK_ANSI_Grave` is located in a different location.
Differential Revision: https://developer.blender.org/D12905
|
|
|
|
The conversion from double to float was causing a build failure.
Differential Revision: https://developer.blender.org/D12946
|
|
|
|
Easy now thanks to the main and shadow path decoupling. Doesn't help
in an benchmark scene except Spring, where it reduces render time by
maybe 2-3%.
Ref T87836
|
|
|
|
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
|
|
Ref T87836
|
|
This patch cleans up code for HIP device and makes it more consistent with the CUDA code.
It also fixes the issue with high VRAM usage on AMD cards using HIP allowing better performance and usage on cards like 6600XT.
Added a check in intern/cycles/kernel/bvh/bvh_util.h to prevent compiler error with hipcc
Reviewed By: brecht, leesonw
Maniphest Tasks: T92124
Differential Revision: https://developer.blender.org/D12834
|
|
|
|
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
|
|
|
|
|
|
Need to initialize components for the full Diffuse BSDF.
Steps to reproduce:
- Default cube scene
- Switch to Cycles renderer
- Enable OSL backend
- Start viewport render
- Observe cube being much black
Differential Revision: https://developer.blender.org/D12921
|
|
A Clang-Format configuration to make the closure definition block to
be properly recognized as such.
Also small wrapper macro to avoid comma in the actual definition code
which was causing unwanted indentation of parameters definition.
Requires Clang-Format 7 or newer. The version we ship in the libs is
12, so for recommended development setup it should all be good.
Differential Revision: https://developer.blender.org/D12920
|
|
The thread affinity setting in OIDN can break multithreading on some CPUs.
While this leads to somewhat worse performance on CPUs that do work correctly,
it's better than having some CPUs use only half the cores.
|
|
Better not to tempt anyone from using unsafe access to device
functionality during host update.
|
|
The lookup tables are to be initialized after device free.
On Linux was only noticeable when rendering default cube scene with
an extra assert. On Windows it was causing an assert in STL in debug
builds.
Differential Revision: https://developer.blender.org/D12918
|
|
|
|
This reverts commit 3065d2609700d14100490a16c91152a6e71790e8. Causing crashes
in the spring scene.
|
|
Ref D12889
|
|
* isect Ng is no longer needed for shadows, for main path needed for SSS only
* Reduce rng_offset and queued_kernel to 16 bits
Ref D12889
|
|
Only copy the number of items used instead of the max items.
Ref D12889
|
|
This is only used by a single device, not need for thread safety.
|
|
|
|
* 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
|
|
|
|
|
|
|
|
Caused a debug crash in Windows MSVS.
Reviewed By: brecht
Differential Revision: https://developer.blender.org/D12873
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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
|
|
The assumption about absent shadow path was wrong.
The rest of the changes are to ensure shadow paths are finished prior
to the split, so that they write to the proper passes.
The issue was caught by running regression tests on OptiX.
Differential Revision: https://developer.blender.org/D12857
|
