diff options
| author | Anthony Barbier <anthony.barbier@arm.com> | 2018-08-08 13:20:04 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:54:54 +0000 |
| commit | b6eb35371d222c6b7f61210d97ebd7dd9e197458 (patch) | |
| tree | af89729ad68d665916c37abb5fd49e512fa40614 /src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp | |
| parent | 1d1f32ce7ef6acea4afd4cf6a929436640b72ccd (diff) | |
| download | ComputeLibrary-b6eb35371d222c6b7f61210d97ebd7dd9e197458.tar.gz | |
COMPMID-1478: Stop relying on static default OpenCL objects in cl2.hpp
This causes problems when ACL is used as a shared library on Android.
Fixes some problems related to creation / destruction order between the Graph's CL backend and core / runtime
Change-Id: I716d63fd42f4586df1ffbb6fa97e4db06d3a781b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143228
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Diffstat (limited to 'src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp')
| -rw-r--r-- | src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp b/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp index 69ab590540..ace3fd5840 100644 --- a/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp +++ b/src/core/CL/kernels/CLConvertFullyConnectedWeightsKernel.cpp @@ -73,7 +73,7 @@ void CLConvertFullyConnectedWeightsKernel::configure(const ICLTensor *input, ICL // Configure kernel window Window win = calculate_max_window(*input->info(), Steps()); - ICLKernel::configure(win); + ICLKernel::configure_internal(win); } Status CLConvertFullyConnectedWeightsKernel::validate(const ITensorInfo *input, const ITensorInfo *output, const TensorShape &original_input_shape, @@ -109,4 +109,4 @@ void CLConvertFullyConnectedWeightsKernel::run(const Window &window, cl::Command add_2D_tensor_argument(idx, _output, window); enqueue(queue, *this, window); } -} // namespace arm_compute
\ No newline at end of file +} // namespace arm_compute |