diff options
author | Sam Laynton <sam.laynton@arm.com> | 2018-04-05 13:26:08 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:49:37 +0000 |
commit | 56e8e8627e5c6912a482a68ec4d65051a5ff4ee8 (patch) | |
tree | 0af347b19d97fb55f8a231c5e7ab373cefc33ea6 /src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp | |
parent | 9fb2b667b8cced30c4585e2aa164b92d8d406676 (diff) | |
download | ComputeLibrary-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/CLGEMMMatrixMultiplyKernel.cpp')
-rw-r--r-- | src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp | 6 |
1 files changed, 5 insertions, 1 deletions
diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp index dc9c59d2d0..0f0646c1ce 100644 --- a/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp +++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyKernel.cpp @@ -219,6 +219,10 @@ void CLGEMMMatrixMultiplyKernel::configure(const ICLTensor *input0, const ICLTen break; case GPUTarget::G71: case GPUTarget::G72: + case GPUTarget::G51: + case GPUTarget::G51BIG: + case GPUTarget::G51LIT: + case GPUTarget::TNOX: if(input1->info()->dimension(1) == 24) { // LWS optimized for the 11x11 AlexNet convolution on Bifrost. @@ -284,7 +288,7 @@ void CLGEMMMatrixMultiplyKernel::configure(const ICLTensor *input0, const ICLTen build_opts.add_option("-DCOLS_A=" + support::cpp11::to_string(input0->info()->dimension(0))); // Create kernels according to the architecture, data type and input size. - if((gpu_target == GPUTarget::G71 || gpu_target == GPUTarget::G72) && data_type == DataType::F32) + if(gpu_target_is_in(gpu_target, GPUTarget::G71, GPUTarget::G72, GPUTarget::G51, GPUTarget::G51BIG, GPUTarget::G51LIT, GPUTarget::TNOX) && data_type == DataType::F32) { // The first kernel is optimized for the case of 1000 or less output elements (e.g. FC8 of AlexNet and VGG-16, and // FC1 of Inception v3). The second kernel is optimized for the case of greater than 1000 output elements (e.g. |