aboutsummaryrefslogtreecommitdiff
path: root/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
diff options
context:
space:
mode:
authorMichele Di Giorgio <michele.digiorgio@arm.com>2019-06-19 12:28:47 +0100
committerMichele Di Giorgio <michele.digiorgio@arm.com>2019-07-03 09:36:46 +0000
commit4646d2e026a0fa92085fcba2f4aec5ec148956aa (patch)
tree0847684b138f9b4631cf56e20d0e75ef313607d0 /src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
parentcbe39055f61bc7a3f6aef4588de47d48b77cf354 (diff)
downloadComputeLibrary-4646d2e026a0fa92085fcba2f4aec5ec148956aa.tar.gz
COMPMID-2072: Use a constexpr for num_elems_processed_per_iteration where possible
Change-Id: I26cb699ae3a77003ef7d05ac30d3ed518214e25f Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/1375 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Manuel Bottini <manuel.bottini@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Pablo Marquez <pablo.tello@arm.com>
Diffstat (limited to 'src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp')
-rw-r--r--src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp15
1 files changed, 7 insertions, 8 deletions
diff --git a/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
index 0ff2f1343a..9e5d677e89 100644
--- a/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixVectorMultiplyKernel.cpp
@@ -33,10 +33,13 @@
#include "arm_compute/core/Helpers.h"
#include "arm_compute/core/Types.h"
-using namespace arm_compute;
-
+namespace arm_compute
+{
namespace
{
+constexpr unsigned int num_elems_read_per_iteration = 4;
+constexpr unsigned int num_rows_read_per_iteration = 4;
+
Status validate_arguments(const ITensorInfo *input0, const ITensorInfo *input1, const ITensorInfo *output)
{
ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(input0);
@@ -50,9 +53,6 @@ Status validate_arguments(const ITensorInfo *input0, const ITensorInfo *input1,
std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input0, ITensorInfo *input1, ITensorInfo *output)
{
- constexpr unsigned int num_elems_read_per_iteration = 4;
- constexpr unsigned int num_rows_read_per_iteration = 4;
-
const unsigned int border_x = ceil_to_multiple(input0->dimension(0), num_elems_read_per_iteration) - input0->dimension(0);
const unsigned int border_y = ceil_to_multiple(input0->dimension(1), num_rows_read_per_iteration) - input0->dimension(1);
@@ -113,9 +113,7 @@ void CLGEMMMatrixVectorMultiplyKernel::configure(const ICLTensor *input0, const
}
// Configure kernel window
- const unsigned int num_elems_read_per_iteration = 4;
-
- _num_rows_read_per_iteration = 4;
+ _num_rows_read_per_iteration = num_rows_read_per_iteration;
const unsigned int border_x = ceil_to_multiple(input0->info()->dimension(0), num_elems_read_per_iteration) - input0->info()->dimension(0);
const unsigned int border_y = ceil_to_multiple(input0->info()->dimension(1), _num_rows_read_per_iteration) - input0->info()->dimension(1);
@@ -172,3 +170,4 @@ void CLGEMMMatrixVectorMultiplyKernel::run(const Window &window, cl::CommandQueu
}
while(window.slide_window_slice_3D(slice_in) && window.slide_window_slice_3D(slice_out));
}
+} // namespace arm_compute