commit | 3502613ec7223a415c6f648d8e0b54d387e00d7d | [log] [tgz] |
---|---|---|
author | Gian Marco <gianmarco.iodice@arm.com> | Thu Feb 22 15:28:44 2018 +0000 |
committer | Anthony Barbier <anthony.barbier@arm.com> | Fri Nov 02 16:47:40 2018 +0000 |
tree | 291f5414a59586c7c4bb99423b21545400c2bfd9 | |
parent | ce876123707373f9a428f43df9d244f1de2c0cbf [diff] [blame] |
COMPMID-765 - Fix performance issues on OpenCL The problem was related to the reshape of the weights. The reshaping happened for each run Change-Id: Ie7d02fa6bb08df34e44213303e9eb0700ff77160 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121877 Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com> Reviewed-by: Pablo Tello <pablo.tello@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
diff --git a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h index 2402950..5f8830a 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h
@@ -176,6 +176,7 @@ CLTensor _tmp_output; bool _is_quantized; + bool _is_first_run; }; } #endif /* __ARM_COMPUTE_CLGEMMCONVOLUTIONLAYER_H__ */