This allows us to do faster native scalar operations. Also
updated half/quarter packets to use the native type if available.
Benchmark improvement:
```
Comparing ./2910_without_float16 to ./2910_with_float16
Benchmark Time CPU Time Old Time New CPU Old CPU New
------------------------------------------------------------------------------------------------------------------------------------
BM_CalcMat<float>/10000/768/500 -0.0041 -0.0040 58276392 58039442 58273420 58039582
BM_CalcMat<_Float16>/10000/768/500 +0.0073 +0.0073 642506339 647214446 642481384 647188303
BM_CalcMat<Eigen::half>/10000/768/500 -0.3170 -0.3170 92511115 63182101 92506771 63179258
BM_CalcVec<float>/10000/768/500 +0.0022 +0.0022 5198157 5209469 5197913 5209334
BM_CalcVec<_Float16>/10000/768/500 +0.0025 +0.0026 10133324 10159111 10132641 10158507
BM_CalcVec<Eigen::half>/10000/768/500 -0.7760 -0.7760 45337937 10156952 45336532 10156389
OVERALL_GEOMEAN -0.2677 -0.2677 0 0 0 0
```
Fixes#2910.
Activates vectorization of the Eigen::half versions of the tanh and
logistic functions when they run on Neon. Both functions convert their
inputs to float before computing the output, and as a result of this
commit, the conversions and the computation in float are vectorized.
The `Complex.h` file applies equally to HIP/CUDA, so placing under the
generic `GPU` folder.
The `TensorReductionCuda.h` has already been deprecated, now removing
for the next Eigen version.
The `Serializer<T>` class implements a binary serialization that
can write to (`serialize`) and read from (`deserialize`) a byte
buffer. Also added convenience routines for serializing
a list of arguments.
This will mainly be for testing, specifically to transfer data to
and from the GPU.
NVCC does not understand `__forceinline`, so we need to use `inline`
when compiling for GPU.
ICC specializes `std::complex` operators for `float` and `double`
by default, which cannot be used on device and conflict with Eigen's
workaround in CUDA/Complex.h. This can be prevented by defining
`_OVERRIDE_COMPLEX_SPECIALIZATION_` before including `<complex>`.
Added this define to the tests and to `Eigen/Core`, but this will
not work if the user includes `<complex>` before `<Eigen/Core>`.
ICC also seems to generate a duplicate `Map` symbol in
`PlainObjectBase`:
```
error: "Map" has already been declared in the current scope
static ConstMapType Map(const Scalar *data)
```
I tracked this down to `friend class Eigen::Map`. Putting the `friend`
statements at the bottom of the class seems to resolve this issue.
Fixes#2180
Clang does a poor job of optimizing the GEBP microkernel on 32-bit ARM,
leading to excessive 16-byte register spills, slowing down basic f32
matrix multiplication by approx 50%.
By specializing `gebp_traits`, we can eliminate the register spills.
Volatile inline ASM both acts as a barrier to prevent reordering and
enforces strict register use. In a simple f32 matrix multiply example,
this modification reduces 16-byte spills from 109 instances to zero,
leading to a 1.5x speed increase (search for `16-byte Spill` in the
assembly in https://godbolt.org/z/chsPbE).
This is a replacement of !379. See there for further discussion.
Also moved `gebp_traits` specializations for NEON to
`Eigen/src/Core/arch/NEON/GeneralBlockPanelKernel.h` to be alongside
other NEON-specific code.
Fixes#2138.
This patch adds support for Arm's new vector extension SVE (Scalable Vector Extension). In contrast to other vector extensions that are supported by Eigen, SVE types are inherently *sizeless*. For the use in Eigen we fix their size at compile-time (note that this is not necessary in general, SVE is *length agnostic*).
During compilation the flag `-msve-vector-bits=N` has to be set where `N` is a power of two in the range of `128`to `2048`, indicating the length of an SVE vector.
Since SVE is rather young, we decided to disable it by default even if it would be available. A user has to enable it explicitly by defining `EIGEN_ARM64_USE_SVE`.
This patch introduces the packet types `PacketXf` and `PacketXi` for packets of `float` and `int32_t` respectively. The size of these packets depends on the SVE vector length. E.g. if `-msve-vector-bits=512` is set, `PacketXf` will contain `512/32 = 16` elements.
This MR is joint work with Miguel Tairum <miguel.tairum@arm.com>.
