diff options
author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2017-07-03 17:41:47 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:15:39 +0100 |
commit | 8a383694445dfebb84732b19d5b3299961e8ffe3 (patch) | |
tree | 09f7521ec6112e7eab12ca2ea74cfbe59ea7d636 /arm_compute/core/CL | |
parent | bdb6b0bb156588dc39fd5084d4c91d05b5148610 (diff) | |
download | ComputeLibrary-8a383694445dfebb84732b19d5b3299961e8ffe3.tar.gz |
COMPMID-434 - Port CLGEMM to support 16 bit fixed point
Change-Id: I30aef3c7ecd1ee740c2a7f2ce65a63c7dcd66e49
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/79630
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Diffstat (limited to 'arm_compute/core/CL')
4 files changed, 4 insertions, 4 deletions
diff --git a/arm_compute/core/CL/kernels/CLGEMMInterleave4x4Kernel.h b/arm_compute/core/CL/kernels/CLGEMMInterleave4x4Kernel.h index 9466b16a91..203e0fc6c4 100644 --- a/arm_compute/core/CL/kernels/CLGEMMInterleave4x4Kernel.h +++ b/arm_compute/core/CL/kernels/CLGEMMInterleave4x4Kernel.h @@ -64,7 +64,7 @@ public: CLGEMMInterleave4x4Kernel &operator=(CLGEMMInterleave4x4Kernel &&) = default; /** Initialise the kernel's input and output. * - * @param[in] input Input tensor. Data types supported: U8/S8/QS8/U16/S16/F16/U32/S32/F32 + * @param[in] input Input tensor. Data types supported: U8/S8/QS8/U16/S16/QS16/F16/U32/S32/F32 * @param[out] output Output tensor. Data type supported: same as @p input */ void configure(const ICLTensor *input, ICLTensor *output); diff --git a/arm_compute/core/CL/kernels/CLGEMMMatrixAdditionKernel.h b/arm_compute/core/CL/kernels/CLGEMMMatrixAdditionKernel.h index b3a85a1706..ada67d1eaf 100644 --- a/arm_compute/core/CL/kernels/CLGEMMMatrixAdditionKernel.h +++ b/arm_compute/core/CL/kernels/CLGEMMMatrixAdditionKernel.h @@ -52,7 +52,7 @@ public: * * @note The input and output tensors must have the same dimensions * - * @param[in] input Input tensor (Matrix C). Data types supported: QS8/F16/F32 + * @param[in] input Input tensor (Matrix C). Data types supported: QS8/QS16/F16/F32 * @param[in, out] output Output tensor. If this kernel is used to finalize the GEMM result (alpha * AB + beta * C), output must contain the result obtained by @ref CLGEMMMatrixMultiplyKernel. Data type supported: same as @p input * @param[in] beta Weight of matrix C */ diff --git a/arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h b/arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h index 7625358b8b..dec63e0679 100644 --- a/arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h +++ b/arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h @@ -53,7 +53,7 @@ public: CLGEMMMatrixMultiplyKernel &operator=(CLGEMMMatrixMultiplyKernel &&) = default; /** Initialise the kernel's input, output and alpha * - * @param[in] input0 Input tensor containing the interleaved Matrix A or the vector A. Data types supported: QS8/F16/F32 + * @param[in] input0 Input tensor containing the interleaved Matrix A or the vector A. Data types supported: QS8/QS16/F16/F32 * @param[in] input1 Input tensor containing the transposed Matrix B if the first input tensor A is not a vector. * If the output tensor is a vector, input1 must contain the matrix B not reshaped. Data type supported: same as @p input0 * @param[out] output Output tensor to store the result of matrix multiplication. Data type supported: same as @p input0 diff --git a/arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h b/arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h index 9657a2af45..0e467aac13 100644 --- a/arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h +++ b/arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h @@ -70,7 +70,7 @@ class CLGEMMTranspose1xWKernel : public ICLSimple2DKernel public: /** Initialise the kernel's input and output. * - * @param[in] input Input tensor. Data types supported: U8/S8/QS8/U16/S16/F16/U32/S32/F32 + * @param[in] input Input tensor. Data types supported: U8/S8/QS8/U16/S16/QS16/F16/U32/S32/F32 * @param[out] output Output tensor. Data type supported: same as @p input */ void configure(const ICLTensor *input, ICLTensor *output); |