commit | 758b5ba3e6d22509d4deab3d8b0b9c2f03418130 | [log] [tgz] |
---|---|---|
author | Matthew Bentham <matthew.bentham@arm.com> | Thu Mar 05 23:37:48 2020 +0000 |
committer | Michele Di Giorgio <michele.digiorgio@arm.com> | Tue Mar 10 18:41:04 2020 +0000 |
tree | c16734e12b105819a919e8015a4d281dffd8817c | |
parent | 6b3865ad038d60a126fe1f90df815a480527a29f [diff] [blame] |
COMPMID-3069: Improve build time by splitting up ToolchainSupport.h Split out the parts of ToolchainSupport coming from <memory> and the parts coming from <string> into their own new header files. This accounts for 99% of uses of ToolchainSupport, which means that expensive header files such as arm_neon.h don't need to be included everywhere. Knocks about 10% of compilation time off kernel files. Signed-off-by: Matthew Bentham <matthew.bentham@arm.com> Change-Id: I2ae718fe766b5ff28608812b0f686f30eeac1b21 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2852 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
diff --git a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel.cpp b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel.cpp index dd1be74..7097dc9 100644 --- a/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel.cpp +++ b/src/core/CL/kernels/CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel.cpp
@@ -33,7 +33,7 @@ #include "arm_compute/core/Window.h" #include "arm_compute/core/utils/misc/ShapeCalculator.h" -#include "support/ToolchainSupport.h" +#include "support/StringSupport.h" namespace arm_compute {