aboutsummaryrefslogtreecommitdiff
path: root/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
diff options
context:
space:
mode:
authorSam Laynton <sam.laynton@arm.com>2018-04-05 13:26:08 +0100
committerAnthony Barbier <anthony.barbier@arm.com>2018-11-02 16:49:37 +0000
commit56e8e8627e5c6912a482a68ec4d65051a5ff4ee8 (patch)
tree0af347b19d97fb55f8a231c5e7ab373cefc33ea6 /src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
parent9fb2b667b8cced30c4585e2aa164b92d8d406676 (diff)
downloadComputeLibrary-56e8e8627e5c6912a482a68ec4d65051a5ff4ee8.tar.gz
COMPMID-1031: Use LWS hints for G51, G51BIG, G51LIT, and TNOX
Change-Id: Ie07d9225faaef778bdcfdcb56ae42ec95962e48d Signed-off-by: Sam Laynton <sam.laynton@arm.com> Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/126735 Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp')
-rw-r--r--src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp4
1 files changed, 2 insertions, 2 deletions
diff --git a/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp b/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
index b5526c4fca..13ee9a1d14 100644
--- a/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
+++ b/src/core/CL/kernels/CLDirectConvolutionLayerKernel.cpp
@@ -134,7 +134,7 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input, ITen
unsigned int num_elems_written_per_iteration_x = 0;
unsigned int num_elems_written_per_iteration_y = 0;
- if(gpu_target_is_in(target, GPUTarget::G71, GPUTarget::G72) && (kernel_size <= 5) && (conv_stride_x == 1) && (conv_stride_y == 1) && (data_type == DataType::F32))
+ if(gpu_target_is_in(target, GPUTarget::G71, GPUTarget::G72, GPUTarget::G51, GPUTarget::G51BIG, GPUTarget::G51LIT, GPUTarget::TNOX) && (kernel_size <= 5) && (conv_stride_x == 1) && (conv_stride_y == 1) && (data_type == DataType::F32))
{
// Configure kernel window
@@ -309,7 +309,7 @@ void CLDirectConvolutionLayerKernel::configure(const ICLTensor *input, const ICL
CLBuildOptions build_options;
build_options.add_option_if(_biases != nullptr, std::string("-DHAS_BIAS"));
- if(gpu_target_is_in(gpu_target, GPUTarget::G71, GPUTarget::G72) && (kernel_size <= 5) && (_conv_stride_x == 1) && (_conv_stride_y == 1) && (data_type == DataType::F32))
+ if(gpu_target_is_in(gpu_target, GPUTarget::G71, GPUTarget::G72, GPUTarget::G51, GPUTarget::G51BIG, GPUTarget::G51LIT, GPUTarget::TNOX) && (kernel_size <= 5) && (_conv_stride_x == 1) && (_conv_stride_y == 1) && (data_type == DataType::F32))
{
build_options.add_option(std::string("-DWEIGHTS_DEPTH=" + support::cpp11::to_string(_weights->info()->dimension(2))));