COMPMID-3684: Use case data type decoupling
Decouples data types for NEFloorKernel
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I6756300540bc5ef32a9990246eed8619a76855f2
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4084
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/core/NEON/kernels/floor/impl/fp16_neon_floor.cpp b/src/core/NEON/kernels/floor/impl/fp16_neon_floor.cpp
new file mode 100644
index 0000000..3222fa9
--- /dev/null
+++ b/src/core/NEON/kernels/floor/impl/fp16_neon_floor.cpp
@@ -0,0 +1,63 @@
+/*
+ * Copyright (c) 2020 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.
+ */
+#if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS)
+
+#include "src/core/NEON/NEMath.h"
+#include "src/core/common/StdTypes.h"
+#include "src/core/common/Validate.h"
+
+#include <arm_neon.h>
+#include <cmath>
+#include <cstddef>
+
+namespace arm_compute
+{
+namespace cpu
+{
+constexpr int step = 8;
+
+void fp16_neon_floor(const void *src, void *dst, int len)
+{
+ ARM_COMPUTE_ASSERT_NOT_NULLPTR(src);
+ ARM_COMPUTE_ASSERT_NOT_NULLPTR(dst);
+ ARM_COMPUTE_ASSERT(len >= 0);
+
+ auto psrc = static_cast<const f16 *>(src);
+ auto pdst = static_cast<f16 *>(dst);
+
+ for(; len >= step; len -= step)
+ {
+ vst1q_f16(pdst, vfloorq_f16(vld1q_f16(psrc)));
+ psrc += step;
+ pdst += step;
+ }
+
+ for(; len >= 0; --len)
+ {
+ *pdst++ = std::floor(*psrc++);
+ }
+}
+} // namespace cpu
+} // namespace arm_compute
+#endif /* defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS) */