Compiler Command: aoc device/vector_add.cl -o bin/vector_add.aocx --report --board c5soc_sharedonly
/home/nicolas/Téléchargements/vector_add/device/vector_add.cl:23:48: warning: declaring kernel argument with no 'restrict' may lead to low kernel performance
__kernel void vector_add(__global const float *x, 
                                               ^
/home/nicolas/Téléchargements/vector_add/device/vector_add.cl:24:48: warning: declaring kernel argument with no 'restrict' may lead to low kernel performance
                         __global const float *y, 
                                               ^
2 warnings generated.
=========================================================================================================
                                   *** Optimization Report ***                                        

Kernels that do not use any work-item built-in functions, such as get_global_id(), are compiled for single
work-item execution (a task). Otherwise, a kernel is compiled as an ND-Range.

For tasks, the compiler will attempt to pipeline every loop in the kernel to allow multiple iterations of
the loop to execute concurrently. If some loops are not pipelined, or not pipelined well, you may not get
good performance.

For ND-Range kernels, the loops are not pipelined. Instead, they are built to accept multiple work-items
simultaneously. Kernel throughput is usually reduced by the largest total number of iterations of nested
loops. A large number of threads is usually required to efficiently utilize ND-Range kernels.

=========================================================================================================
Kernel: vector_add                                                                                       
=========================================================================================================
The kernel is compiled as an ND-Range.

The kernel does not use any work-group information (such as get_local_id() or get_group_id()).
Local work-group size will be automatically modified to match global work-group size on launch.
This is a hardware optimization.

=========================================================================================================

+--------------------------------------------------------------------+
; Estimated Resource Usage Summary                                   ;
+----------------------------------------+---------------------------+
; Resource                               + Usage                     ;
+----------------------------------------+---------------------------+
; Logic utilization                      ;   18%                     ;
; ALUTs                                  ;   11%                     ;
; Dedicated logic registers              ;    7%                     ;
; Memory blocks                          ;   12%                     ;
; DSP blocks                             ;    0%                     ;
+----------------------------------------+---------------------------;
system_integrator: main.cpp:474: int ReadKernelData(altera_xercesc_2_6::XERCES_DOMDocument*, OpenCLSystem*): Assertion `document != __null' failed.
Aborted (core dumped)