diff options
-rw-r--r-- | src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp | 2 | ||||
-rw-r--r-- | src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp index 85d882c865..966bca099a 100644 --- a/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp +++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp @@ -133,7 +133,7 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input0, ITe // Note: bottom paddings are calculated manually as the output can be reinterpreted as 3D tensor // The only way to set properly the paddings, it is to set those explicitly through the AccessWindowStatic - const int m = gemm_info.m(); + const int m = reinterpret_output_as_3d ? gemm_info.m() : input0->dimension(1); const int bottom_pad = (num_elems_processed_per_iteration_y - (m % num_elems_processed_per_iteration_y)) % num_elems_processed_per_iteration_y; win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y)); diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp index 58c4cdd2f6..f764b9d36d 100644 --- a/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp +++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyReshapedOnlyRHSKernel.cpp @@ -154,7 +154,7 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input0, ITe // Note: bottom paddings are calculated manually as the output can be reinterpreted as 3D tensor // The only way to set properly the paddings, it is to set those explicitly through the AccessWindowStatic - const int m = gemm_info.m(); + const int m = reinterpret_output_as_3d ? gemm_info.m() : input0->dimension(1); const int bottom_pad = (num_elems_processed_per_iteration_y - (m % num_elems_processed_per_iteration_y)) % num_elems_processed_per_iteration_y; win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y)); |