This PR allows to run benchmarks on multiple GPUs (one at a time).
It also contains some fixes to the reporting and normalization of the
device ID's.
Device ID normalization is done as some existing configs only contain
`VULKAN` which need to be matched against `VULKAN_0` and vise-versa.
Device selection doesn't use `blender --gpu-device help` to detect devices
as that option doesn't exist in older builds.
Now with fixed python validations.
Original Pull Request: https://projects.blender.org/blender/blender/pulls/158898
Pull Request: https://projects.blender.org/blender/blender/pulls/159234
This PR allows to run benchmarks on multiple GPUs (one at a time).
It also contains some fixes to the reporting and normalization of the
device ID's.
Device ID normalization is done as some existing configs only contain
`VULKAN` which need to be matched against `VULKAN_0` and vise-versa.
Device selection doesn't use `blender --gpu-device help` to detect devices
as that option doesn't exist in older builds.
Pull Request: https://projects.blender.org/blender/blender/pulls/158898
Adding the actual GPU device name and host CPU to the test results.
For Cycles the CPU isn't that important when using GPUs as it would
mostly perform the scheduling, however when running EEVEE tests the
CPU is more involved.
This PR will extract the correct GPU device names and adds the CPU name
to a separate identifier. The used CPU is then added to the report, when
found in the results and different than the test device.
Pull Request: https://projects.blender.org/blender/blender/pulls/155419
Adding support to use multiple GPU backends when running EEVEE
performance tests.
The backends to test can be configured in the devices property of the
config. When not explicitly configured it will run with 'CPU' device.
```python
devices = ['CPU', 'OPENGL', 'VULKAN']
tests = ['*']
categories = ['eevee']
revisions = {
'v5.2': 'b9d4616281',
'v5.1': '34ce889d74',
}
```
Every test can override `supported_device_types` to tell the runner
for which device type it should run the test.
This change will also remove the `time` field from the report as it
isn't useful for EEVEE viewport tests. It was initially added to be
compatible with the Cycles test.
See PR for example reports and output.
Pull Request: https://projects.blender.org/blender/blender/pulls/154117
This commit gives users of the Cycles performance benchmark tool the
option to run performance benchmarks with OSL enabled for CPUs
and OptiX devices.
This can be done by adding `-OSL` to the device name:
`CPU-OSL`
`OPTIX-OSL_0`
Pull Request: https://projects.blender.org/blender/blender/pulls/136506
Similar to the regression tests it is possible to append -RT suffix
to the compute type to enable hardware ray-tracing.
For example HIP_0 will run benchmark on first HIP device without
hardware ray-tracing, HIP-RT_0 will run benchmark on the same device
but will enable hardware ray-tracing.
The downside of this change is that it will make it so METAL device
will no longer use HW-RT on M3 and aboce, and explicit METAL-RT is
to be used. This is because benchmark was relying on the Auto
configuration which has different behavior depending on the device
generation.
Pull Request: https://projects.blender.org/blender/blender/pulls/136308
Listing the "Blender Foundation" as copyright holder implied the Blender
Foundation holds copyright to files which may include work from many
developers.
While keeping copyright on headers makes sense for isolated libraries,
Blender's own code may be refactored or moved between files in a way
that makes the per file copyright holders less meaningful.
Copyright references to the "Blender Foundation" have been replaced with
"Blender Authors", with the exception of `./extern/` since these this
contains libraries which are more isolated, any changed to license
headers there can be handled on a case-by-case basis.
Some directories in `./intern/` have also been excluded:
- `./intern/cycles/` it's own `AUTHORS` file is planned.
- `./intern/opensubdiv/`.
An "AUTHORS" file has been added, using the chromium projects authors
file as a template.
Design task: #110784
Ref !110783.
Use a shorter/simpler license convention, stops the header taking so
much space.
Follow the SPDX license specification: https://spdx.org/licenses
- C/C++/objc/objc++
- Python
- Shell Scripts
- CMake, GNUmakefile
While most of the source tree has been included
- `./extern/` was left out.
- `./intern/cycles` & `./intern/atomic` are also excluded because they
use different header conventions.
doc/license/SPDX-license-identifiers.txt has been added to list SPDX all
used identifiers.
See P2788 for the script that automated these edits.
Reviewed By: brecht, mont29, sergey
Ref D14069
These are scripts for benchmarking Blender features on real-world .blend
files. They were originally written for benchmarking Cycles performance, and
were made generic so they can be used for more Blender features.
The benchmarks can be run locally by developers. But the plan is to also run
these as part of continuous integration to track performance over time.
Currently there are tests for Cycles rendering and .blend file loading.
Documentation:
https://wiki.blender.org/wiki/Tools/Tests/Performance
Main features:
* User created configurations to quickly run, re-run and analyze a selected
subset of tests.
* Supports both benchmarking with existing builds, and automatic building of
specified git commits, tags and branches.
* Generate HTML page with bar and line graphs from test results.
* Controlled using simple command line tool.
* For writing tests, convenient abstraction to run a Python function in Blender
with arguments and return value.
Ref T74730
Differential Revision: https://developer.blender.org/D11662
