diff options
| author | ASIAPAC\steli01 <stephen.li@arm.com> | 2017-11-07 16:14:44 +0800 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:42:17 +0000 |
| commit | 23ac91b6ba235e67847802d4b49e494fa5bedbb6 (patch) | |
| tree | 49824ba9d389ee37a7656c10b960aaac5aba3567 /src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp | |
| parent | 923241eb998ad031f4cce7b12d8c24a0b6c80be8 (diff) | |
| download | ComputeLibrary-23ac91b6ba235e67847802d4b49e494fa5bedbb6.tar.gz | |
APPBROWSER-290: DC5x5 optimization for FP16
Change-Id: I0833437d8353515a29aad69c15c99c7b0fd65156
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/111631
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Joel Liang <joel.liang@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Diffstat (limited to 'src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp')
| -rw-r--r-- | src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp b/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp index 962f044214..b032bc5668 100644 --- a/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp +++ b/src/core/GLES_COMPUTE/kernels/GCDirectConvolutionLayerKernel.cpp @@ -211,6 +211,7 @@ void GCDirectConvolutionLayerKernel<kernel_size>::configure(const IGCTensor *inp switch(input->info()->data_type()) { case DataType::F16: + options.emplace("#define PROCESS_4X_1Y_1Z"); num_elems_read_per_iteration_x = 8; num_elems_written_per_iteration_x = 4; |