diff options
author | SiCong Li <sicong.li@arm.com> | 2020-11-08 21:58:01 +0000 |
---|---|---|
committer | SiCong Li <sicong.li@arm.com> | 2020-11-09 10:21:44 +0000 |
commit | 18bdfaefd3aeb2557c2e484b67b06c515f9a120f (patch) | |
tree | 82451702f8651445daa8ab0d91a52d9b0c10339d /src/core/CL/cl_kernels/space_to_batch.cl | |
parent | 0ea50e3233c0b20a3e3c68e42bdff31565cefa3d (diff) | |
download | ComputeLibrary-18bdfaefd3aeb2557c2e484b67b06c515f9a120f.tar.gz |
COMPMID-3951 LargeGraph_FLOAT32_Rank4_25 CTS failures in Android Q in CL Fix1
* Fix CLSpaceToBatchLayerKernel and NESpaceToBatchLayerKernel validation
errors by using the correctly calculated output tensor shape
Signed-off-by: SiCong Li <sicong.li@arm.com>
Change-Id: I21d61f870e6a23a2e38dcb95c939b0bf08082b6f
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4347
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: TeresaARM <teresa.charlinreyes@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/CL/cl_kernels/space_to_batch.cl')
-rw-r--r-- | src/core/CL/cl_kernels/space_to_batch.cl | 4 |
1 files changed, 0 insertions, 4 deletions
diff --git a/src/core/CL/cl_kernels/space_to_batch.cl b/src/core/CL/cl_kernels/space_to_batch.cl index 5ade9c5a7c..cb11786ac4 100644 --- a/src/core/CL/cl_kernels/space_to_batch.cl +++ b/src/core/CL/cl_kernels/space_to_batch.cl @@ -46,8 +46,6 @@ * @param[in] block_shape_ptr Pointer to the block shape tensor. Supported data types: S32 * @param[in] block_shape_stride_x Stride of the block shape tensor in X dimension (in bytes) * @param[in] block_shape_step_x block_shape_stride_x * number of elements along X processed per workitem(in bytes) - * @param[in] block_shape_stride_y Stride of the block shape tensor in Y dimension (in bytes) - * @param[in] block_shape_step_y block_shape_stride_y * number of elements along Y processed per workitem(in bytes) * @param[in] block_shape_offset_first_element_in_bytes The offset of the first element in the block shapetensor * @param[in] batch_id The output tensor batch id * @param[out] output_ptr Pointer to the destination tensor. Supported data types: same as @p input_ptr @@ -117,8 +115,6 @@ __kernel void space_to_batch_nchw( * @param[in] block_shape_ptr Pointer to the block shape tensor. Supported data types: S32 * @param[in] block_shape_stride_x Stride of the block shape tensor in X dimension (in bytes) * @param[in] block_shape_step_x block_shape_stride_x * number of elements along X processed per workitem(in bytes) - * @param[in] block_shape_stride_y Stride of the block shape tensor in Y dimension (in bytes) - * @param[in] block_shape_step_y block_shape_stride_y * number of elements along Y processed per workitem(in bytes) * @param[in] block_shape_offset_first_element_in_bytes The offset of the first element in the block shapetensor * @param[in] batch_id The output tensor batch id * @param[out] output_ptr Pointer to the destination tensor. Supported data types: same as @p input_ptr |