aboutsummaryrefslogtreecommitdiff
path: root/src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp
diff options
context:
space:
mode:
authorGeorgios Pinitas <georgios.pinitas@arm.com>2018-06-04 19:27:13 +0100
committerAnthony Barbier <anthony.barbier@arm.com>2018-11-02 16:52:54 +0000
commit17812ba9f7cf2c8f5121c11760ac45fbbdb7aeaf (patch)
tree28c7bb65a8306e82de91a644fdcc1c0947c6f6d7 /src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp
parentf8d8f3aff04faf731f20411ecb91027eab4365c5 (diff)
downloadComputeLibrary-17812ba9f7cf2c8f5121c11760ac45fbbdb7aeaf.tar.gz
COMPMID-817: Tuner: Port kernels to new design.
Change-Id: Iaabb1153c2abe0400ec79d51a21347debe92d642 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/134062 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Diffstat (limited to 'src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp')
-rw-r--r--src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp2
1 files changed, 2 insertions, 0 deletions
diff --git a/src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp b/src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp
index 676a121a76..c2b24e3c20 100644
--- a/src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp
+++ b/src/runtime/CL/functions/CLDepthwiseConvolutionLayer.cpp
@@ -134,6 +134,7 @@ void CLDepthwiseConvolutionLayer::configure(ICLTensor *input, const ICLTensor *w
_input_reshaped.allocator()->init(input->info()->clone()->set_is_resizable(true).reset_padding().set_tensor_shape(shape_im2col));
_im2col_kernel.set_target(gpu_target);
_im2col_kernel.configure(input, &_input_reshaped, Size2D(weights_w, weights_h), conv_info, append_bias, depth_multiplier);
+ CLScheduler::get().tune_kernel_static(_im2col_kernel);
// Weights reshape configuration
const TensorShape shape_weights_reshape(patch_size, weights_z);
@@ -149,6 +150,7 @@ void CLDepthwiseConvolutionLayer::configure(ICLTensor *input, const ICLTensor *w
_v2mm_output.allocator()->init(input->info()->clone()->set_is_resizable(true).reset_padding().set_data_type(v2mm_dt).set_tensor_shape(shape_v2mm_out));
_v2mm_kernel.set_target(gpu_target);
_v2mm_kernel.configure(&_input_reshaped, &_weights_reshaped, &_v2mm_output);
+ CLScheduler::get().tune_kernel_static(_v2mm_kernel);
_output_reshaped.allocator()->init(_v2mm_output.info()->clone()->set_is_resizable(true).reset_padding().set_tensor_shape(output_shape));
_vector_to_tensor_kernel.configure(&_v2mm_output, (_is_quantized) ? &_output_reshaped : output, conv_w, conv_h);