We used it only to access device id for explicitly allowing Arc GPUs.
It made the backend require ze_loader.dll which could be problematic if
we end up using direct linking.
I've replaced filtering based on PCI device id by using other HW properties
instead (EUs, threads per EU), that are now available through Level-Zero.
Initially oneAPI implementation have waited after each memory
operation, even if there was no need for this. Now, the implementation
will wait only if it is really necessary - it have improved
performance noticeble for some scenes and a bit for the rest of them.
This patch adds a new Cycles device with similar functionality to the
existing GPU devices. Kernel compilation and runtime interaction happen
via oneAPI DPC++ compiler and SYCL API.
This implementation is primarly focusing on Intel® Arc™ GPUs and other
future Intel GPUs. The first supported drivers are 101.1660 on Windows
and 22.10.22597 on Linux.
The necessary tools for compilation are:
- A SYCL compiler such as oneAPI DPC++ compiler or
https://github.com/intel/llvm
- Intel® oneAPI Level Zero which is used for low level device queries:
https://github.com/oneapi-src/level-zero
- To optionally generate prebuilt graphics binaries: Intel® Graphics
Compiler All are included in Linux precompiled libraries on svn:
https://svn.blender.org/svnroot/bf-blender/trunk/lib The same goes for
Windows precompiled binaries but for the graphics compiler, available
as "Intel® Graphics Offline Compiler for OpenCL™ Code" from
https://www.intel.com/content/www/us/en/developer/articles/tool/oneapi-standalone-components.html,
for which path can be set as OCLOC_INSTALL_DIR.
Being based on the open SYCL standard, this implementation could also be
extended to run on other compatible non-Intel hardware in the future.
Reviewed By: sergey, brecht
Differential Revision: https://developer.blender.org/D15254
Co-authored-by: Nikita Sirgienko <nikita.sirgienko@intel.com>
Co-authored-by: Stefan Werner <stefan.werner@intel.com>
This could result in wrong skipping of SVM nodes in the graph. Now make the
logic consistent with the clamping in the OSL implementation and constant
folding.
Thanks to Christophe Hery for finding the problem and providing the fix.
Enables Vega and Vega II GPUs as well as Vega APU, using changes in HIP code
to support 64-bit waves and a new HIP SDK version.
Tested with Radeon WX9100, Radeon VII GPUs and Ryzen 7 PRO 5850U with Radeon
Graphics APU.
Ref T96740, T91571
Differential Revision: https://developer.blender.org/D15242
This change helps decrease Intel GPU binaries compile time by 5-10
minutes without impacting other backends.
Reviewed By: sergey, brecht
Differential Revision: http://developer.blender.org/D15273
This patch unifies the names of math functions for different data types and uses
overloading instead. The goal is to make it possible to swap out all the float3
variables containing RGB data with something else, with as few as possible
changes to the code. It's a requirement for future spectral rendering patches.
Differential Revision: https://developer.blender.org/D15276
* Rename "texture" to "data array". This has not used textures for a long time,
there are just global memory arrays now. (On old CUDA GPUs there was a cache
for textures but not global memory, so we used to put all data in textures.)
* For CUDA and HIP, put globals in KernelParams struct like other devices.
* Drop __ prefix for data array names, no possibility for naming conflict now that
these are in a struct.
Move MNEE to own kernel, separate from shader ray-tracing. This does introduce
the limitation that a shader can't use both MNEE and AO/bevel, but that seems
like the better trade-off for now.
We can experiment with bigger kernel organization changes later.
Differential Revision: https://developer.blender.org/D15070
Regenerate blackbody RGB curve fit to not clamp values, and extend down to
800K since it does now change below 965K.
Note that as before, blackbody is only defined in the range 800K to 12000K
and has a fixed value outside of that. But within that range there should
be no more unnecessary gamut clamping.
This patch makes it possible to change the precision with which to
store volume data in the NanoVDB data structure (as float, half, or
using variable bit quantization) via the previously unused precision
field in the volume data block.
It makes it possible to further reduce memory usage during
rendering, at a slight cost to the visual detail of a volume.
Differential Revision: https://developer.blender.org/D10023
Found those missing casts while looking into a crash report made in
the Blender Chat. Was unable to reproduce the crash, but the casts
should totally be there to avoid integer overflow.
Ensure the correct total/diffuse/transmission depth is set when evaluating
shaders for MNEE, consistent with regular light shader evaluation.
Differential Revision: https://developer.blender.org/D14902
Made tangent frame consistent across the surface regardless of the sample,
which was not the case with the previous algorithm. Previously, a tangent
frame would stay consistent for the same sample throughout the walk, but not
from sample to sample for the same triangle. This actually resulted in code
simplification.
Also includes additional fixes:
* Fixed an important bug that manifested itself with multiple lights in the
scene, where caustics had abnormally low amplitude: The final light pdf did
not include the light distribution pdf.
* Removed unnecessary orthonormal basis generation function, using cycles'
native one instead.
* Increased solver max iteration back to 64: It turns out we sometimes need
these extra iterations in cases where projection back to the surface takes
many steps. The effective solver iteration count, the most expensive part,
is actually much less than the raw iteration count.
Differential Revision: https://developer.blender.org/D14931
This completes support for tiled texture packing on the Blender / Cycles
side of things.
Most of these changes fall into one of three categories:
- Updating Image handling code to pack/unpack tiled and multi-view images
- Updating Cycles to handle tiled textures through BlenderImageLoader
- Updating OSL to properly handle textures with multiple slots
Differential Revision: https://developer.blender.org/D14395
It was wrongly writing passes twice, for both the surface entry and exit points.
We can skip code for filtering closures, emission and holdout also, as these do
nothing with only a subsurface diffuse closure present.
