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.
A lot of files were missing copyright field in the header and
the Blender Foundation contributed to them in a sense of bug
fixing and general maintenance.
This change makes it explicit that those files are at least
partially copyrighted by the Blender Foundation.
Note that this does not make it so the Blender Foundation is
the only holder of the copyright in those files, and developers
who do not have a signed contract with the foundation still
hold the copyright as well.
Another aspect of this change is using SPDX format for the
header. We already used it for the license specification,
and now we state it for the copyright as well, following the
FAQ:
https://reuse.software/faq/
Goals of this refactor:
* Reduce memory consumption of `IndexMask`. The old `IndexMask` uses an
`int64_t` for each index which is more than necessary in pretty much all
practical cases currently. Using `int32_t` might still become limiting
in the future in case we use this to index e.g. byte buffers larger than
a few gigabytes. We also don't want to template `IndexMask`, because
that would cause a split in the "ecosystem", or everything would have to
be implemented twice or templated.
* Allow for more multi-threading. The old `IndexMask` contains a single
array. This is generally good but has the problem that it is hard to fill
from multiple-threads when the final size is not known from the beginning.
This is commonly the case when e.g. converting an array of bool to an
index mask. Currently, this kind of code only runs on a single thread.
* Allow for efficient set operations like join, intersect and difference.
It should be possible to multi-thread those operations.
* It should be possible to iterate over an `IndexMask` very efficiently.
The most important part of that is to avoid all memory access when iterating
over continuous ranges. For some core nodes (e.g. math nodes), we generate
optimized code for the cases of irregular index masks and simple index ranges.
To achieve these goals, a few compromises had to made:
* Slicing of the mask (at specific indices) and random element access is
`O(log #indices)` now, but with a low constant factor. It should be possible
to split a mask into n approximately equally sized parts in `O(n)` though,
making the time per split `O(1)`.
* Using range-based for loops does not work well when iterating over a nested
data structure like the new `IndexMask`. Therefor, `foreach_*` functions with
callbacks have to be used. To avoid extra code complexity at the call site,
the `foreach_*` methods support multi-threading out of the box.
The new data structure splits an `IndexMask` into an arbitrary number of ordered
`IndexMaskSegment`. Each segment can contain at most `2^14 = 16384` indices. The
indices within a segment are stored as `int16_t`. Each segment has an additional
`int64_t` offset which allows storing arbitrary `int64_t` indices. This approach
has the main benefits that segments can be processed/constructed individually on
multiple threads without a serial bottleneck. Also it reduces the memory
requirements significantly.
For more details see comments in `BLI_index_mask.hh`.
I did a few tests to verify that the data structure generally improves
performance and does not cause regressions:
* Our field evaluation benchmarks take about as much as before. This is to be
expected because we already made sure that e.g. add node evaluation is
vectorized. The important thing here is to check that changes to the way we
iterate over the indices still allows for auto-vectorization.
* Memory usage by a mask is about 1/4 of what it was before in the average case.
That's mainly caused by the switch from `int64_t` to `int16_t` for indices.
In the worst case, the memory requirements can be larger when there are many
indices that are very far away. However, when they are far away from each other,
that indicates that there aren't many indices in total. In common cases, memory
usage can be way lower than 1/4 of before, because sub-ranges use static memory.
* For some more specific numbers I benchmarked `IndexMask::from_bools` in
`index_mask_from_selection` on 10.000.000 elements at various probabilities for
`true` at every index:
```
Probability Old New
0 4.6 ms 0.8 ms
0.001 5.1 ms 1.3 ms
0.2 8.4 ms 1.8 ms
0.5 15.3 ms 3.0 ms
0.8 20.1 ms 3.0 ms
0.999 25.1 ms 1.7 ms
1 13.5 ms 1.1 ms
```
Pull Request: https://projects.blender.org/blender/blender/pulls/104629
This can improve performance in some circumstances when there are
vectorized and/or unrolled loops. I especially noticed that this helps
a lot while working on D16970 (got a 10-20% speedup there by avoiding
running into the non-vectorized fallback loop too often).
Previously, `ParamsBuilder` lazily allocated an array for an
output when it was unused, but the called multi-function
wanted to access it. Now, whether the multi-function supports
an output to be unused is part of the signature. This way, the
allocation can happen earlier when the parameters are build.
The benefit is that this makes all methods of `MFParams`
thread-safe again, removing the need for a mutex.
This moves all multi-function related code in the `functions` module
into a new `multi_function` namespace. This is similar to how there
is a `lazy_function` namespace.
The main benefit of this is that many types names that were prefixed
with `MF` (for "multi function") can be simplified.
There is also a common shorthand for the `multi_function` namespace: `mf`.
This is also similar to lazy-functions where the shortened namespace
is called `lf`.
Goals:
* Better high level control over where devirtualization occurs. There is always
a trade-off between performance and compile-time/binary-size.
* Simplify using array devirtualization.
* Better performance for cases where devirtualization wasn't used before.
Many geometry nodes accept fields as inputs. Internally, that means that the
execution functions have to accept so called "virtual arrays" as inputs. Those
can be e.g. actual arrays, just single values, or lazily computed arrays.
Due to these different possible virtual arrays implementations, access to
individual elements is slower than it would be if everything was just a normal
array (access does through a virtual function call). For more complex execution
functions, this overhead does not matter, but for small functions (like a simple
addition) it very much does. The virtual function call also prevents the compiler
from doing some optimizations (e.g. loop unrolling and inserting simd instructions).
The solution is to "devirtualize" the virtual arrays for small functions where the
overhead is measurable. Essentially, the function is generated many times with
different array types as input. Then there is a run-time dispatch that calls the
best implementation. We have been doing devirtualization in e.g. math nodes
for a long time already. This patch just generalizes the concept and makes it
easier to control. It also makes it easier to investigate the different trade-offs
when it comes to devirtualization.
Nodes that we've optimized using devirtualization before didn't get a speedup.
However, a couple of nodes are using devirtualization now, that didn't before.
Those got a 2-4x speedup in common cases.
* Map Range
* Random Value
* Switch
* Combine XYZ
Differential Revision: https://developer.blender.org/D14628
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
Previously, there was a fixed grain size for all multi-functions. That was
not sufficient because some functions could benefit a lot from smaller
grain sizes.
This refactors adds a new `MultiFunction::call_auto` method which has the
same effect as just calling `MultiFunction::call` but additionally figures
out how to execute the specific multi-function efficiently. It determines
a good grain size and decides whether the mask indices should be shifted
or not.
Most multi-function evaluations benefit from this, but medium sized work
loads (1000 - 50000 elements) benefit from it the most. Especially when
expensive multi-functions (e.g. noise) is involved. This is because for
smaller work loads, threading is rarely used and for larger work loads
threading worked fine before already.
With this patch, multi-functions can specify execution hints, that allow
the caller to execute it most efficiently. These execution hints still
have to be added to more functions.
Some performance measurements of a field evaluation involving noise and
math nodes, ordered by the number of elements being evaluated:
```
1,000,000: 133 ms -> 120 ms
100,000: 30 ms -> 18 ms
10,000: 20 ms -> 2.7 ms
1,000: 4 ms -> 0.5 ms
100: 0.5 ms -> 0.4 ms
```
Previously, the function names were stored in `std::string` and were often
created dynamically (especially when the function just output a constant).
This resulted in a lot of overhead.
Now the function name is just a `const char *` that should be statically
allocated. This is good enough for the majority of cases. If a multi-function
needs a more dynamic name, it can override the `MultiFunction::debug_name`
method.
In my test file with >400,000 simple math nodes, the execution time improves from
3s to 1s.
Previously, the signature of a `MultiFunction` was always embedded into the function.
There are two issues with that. First, `MFSignature` is relatively large, because it contains
multiple strings and vectors. Secondly, constructing it can add overhead that should not
be necessary, because often the same signature can be reused.
The solution is to only keep a pointer to a signature in `MultiFunction` that is set during
construction. Child classes are responsible for making sure that the signature lives
long enough. In most cases, the signature is either embedded into the child class or
it is allocated statically (and is only created once).
