diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2018-09-04 12:18:50 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:54:54 +0000 |
commit | a34286ecabf4fc9e66e423332063a3d5fb17b8f8 (patch) | |
tree | 849999d27a1b0e67b2980730fce04c753a7b37fe /src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp | |
parent | 71ecf396bb08e302dc06b2c7ed44001894d3a2ea (diff) | |
download | ComputeLibrary-a34286ecabf4fc9e66e423332063a3d5fb17b8f8.tar.gz |
COMPMID-1553: Add GPU detection for G52.
Change-Id: Ief61733d0e66fec0c67255ad899c76933109b797
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/146794
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Diffstat (limited to 'src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp')
-rw-r--r-- | src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp b/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp index 6de97d40af..c3d514adb4 100644 --- a/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp +++ b/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp @@ -93,8 +93,14 @@ Status validate_arguments(const ITensorInfo *input, const ITensorInfo *weights, inline bool can_run_optimized_kernel_for_bifrost(GPUTarget gpu_target, unsigned int conv_stride_x, unsigned int conv_stride_y, unsigned int kernel_size, DataType data_type, DataLayout data_layout) { - return gpu_target_is_in(gpu_target, GPUTarget::G71, GPUTarget::G72, GPUTarget::G51, GPUTarget::G51BIG, GPUTarget::G51LIT, GPUTarget::G76) && (kernel_size <= 5) - && (conv_stride_x == 1) && (conv_stride_y == 1) && (data_type == DataType::F32) && (data_layout == DataLayout::NCHW); + return gpu_target_is_in(gpu_target, + GPUTarget::G71, GPUTarget::G72, GPUTarget::G76, + GPUTarget::G51, GPUTarget::G51BIG, GPUTarget::G51LIT, + GPUTarget::G52, GPUTarget::G52LIT) + && (kernel_size <= 5) + && (conv_stride_x == 1) && (conv_stride_y == 1) + && (data_type == DataType::F32) + && (data_layout == DataLayout::NCHW); } inline void setup_num_elems(unsigned int &num_elems_read_per_iteration_x, unsigned int &num_elems_read_per_iteration_y, |