Implement MatMul Function and Operator with Floating Point support for CPU

- Implements MatMul function and operator for floating point datatype FP16/FP32
- Includes support for transposing dynamic tensors prior to matrix multiplication.
- Adds tests for 2D/3D/4D+ tensors in MatMul with F32/F16 datatype (with all combinations of transposed/not-transposed tensors)
- Updates fixture to allow for testing fused activation in MatMul
- Adds tests for matmul with and without fused activation

Resolved: [COMPMID-5898]
Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com>
Change-Id: Iefa84b26dd723c9a51e6c3f91023152c6c31ace2
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9411
Reviewed-by: SiCong Li <sicong.li@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 75 insertions, 0 deletions

diff --git a/src/runtime/NEON/functions/NEMatMul.cpp b/src/runtime/NEON/functions/NEMatMul.cpp
new file mode 100644
index 0000000000..0c46516f1e
--- /dev/null
+++ b/src/runtime/NEON/functions/NEMatMul.cpp
@@ -0,0 +1,75 @@
+/*
+ * Copyright (c) 2023 Arm Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include "arm_compute/runtime/NEON/functions/NEMatMul.h"
+
+#include "arm_compute/core/Validate.h"
+#include "arm_compute/runtime/MemoryGroup.h"
+#include "src/core/helpers/MemoryHelpers.h"
+#include "src/cpu/operators/CpuMatMul.h"
+#include "arm_compute/runtime/Tensor.h"
+
+namespace arm_compute
+{
+struct NEMatMul::Impl
+{
+    const ITensor                  *lhs{ nullptr };
+    const ITensor                  *rhs{ nullptr };
+    ITensor                        *output{ nullptr };
+    std::unique_ptr<cpu::CpuMatMul> op{ nullptr };
+    MemoryGroup                     memory_group{};
+    WorkspaceData<Tensor>           workspace_tensors{};
+    ITensorPack                     run_pack{};
+};
+
+NEMatMul::NEMatMul()
+    : _impl(std::make_unique<Impl>())
+{
+}
+
+NEMatMul::~NEMatMul() = default;
+
+void NEMatMul::configure(ITensor *lhs, ITensor *rhs, ITensor *output, const MatMulInfo &info, const CpuMatMulSettings &settings)
+{
+    _impl->lhs    = lhs;
+    _impl->rhs    = rhs;
+    _impl->output = output;
+
+    ARM_COMPUTE_ERROR_ON_NULLPTR(_impl->lhs, _impl->rhs, _impl->output);
+    _impl->op = std::make_unique<cpu::CpuMatMul>();
+    _impl->op->configure(lhs->info(), rhs->info(), output->info(), info, settings);
+    _impl->run_pack          = { { ACL_SRC_0, lhs }, { ACL_SRC_1, rhs }, { ACL_DST, output } };
+    _impl->workspace_tensors = manage_workspace<Tensor>(_impl->op->workspace(), _impl->memory_group, _impl->run_pack);
+}
+
+Status NEMatMul::validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &info, const CpuMatMulSettings &settings)
+{
+    return cpu::CpuMatMul::validate(lhs, rhs, output, info, settings);
+}
+
+void NEMatMul::run()
+{
+    MemoryGroupResourceScope scope_mg(_impl->memory_group);
+    _impl->op->run(_impl->run_pack);
+}
+} // namespace arm_compute