diff options
| author | Giorgio Arena <giorgio.arena@arm.com> | 2021-08-20 15:24:03 +0100 |
|---|---|---|
| committer | Giorgio Arena <giorgio.arena@arm.com> | 2021-08-24 11:46:13 +0000 |
| commit | 9c67d3883adaa8a65e3567579cf99937deaab21a (patch) | |
| tree | e3b11c5918b4121e8e7d5ad728847ce74a8ed77f /src/graph/backends/CL/CLDeviceBackend.cpp | |
| parent | 87a74effff65f6fa1b0e565818e02c3b414ae1cf (diff) | |
| download | ComputeLibrary-9c67d3883adaa8a65e3567579cf99937deaab21a.tar.gz | |
Remove map/unmap overhead for input/output accessor when using DummyAccessor
Don't map/unmap when the tensor data is not accessed in the input or output accessor.
This is so to avoid measuring CPU overhead when benchmarking on the GPU backend.
Resolve COMPMID-4712
Change-Id: I5baba1b93e7a51fe13525bcce6c0cfdecb14493e
Signed-off-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/6140
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/graph/backends/CL/CLDeviceBackend.cpp')
| -rw-r--r-- | src/graph/backends/CL/CLDeviceBackend.cpp | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/src/graph/backends/CL/CLDeviceBackend.cpp b/src/graph/backends/CL/CLDeviceBackend.cpp index b6b25cc7d0..01e5ab1730 100644 --- a/src/graph/backends/CL/CLDeviceBackend.cpp +++ b/src/graph/backends/CL/CLDeviceBackend.cpp @@ -221,6 +221,11 @@ std::shared_ptr<arm_compute::IWeightsManager> CLDeviceBackend::create_weights_ma auto weights_mgr = std::make_shared<IWeightsManager>(); return weights_mgr; } + +void CLDeviceBackend::sync() +{ + CLScheduler::get().sync(); +} } // namespace backends } // namespace graph } // namespace arm_compute |