OpenCL* on Intel® Programmable Acceleration Card with Intel® Arria® 10 GX FPGA Quick Start User Guide

Download PDF
ID 683831
Date 6/05/2020
Public
Document Table of Contents
Document Table of Contents x
1. About this Document 2. Setting Up the Host Machine 3. Running Diagnostics 4. OpenCL* Support for Multi-Card Systems 5. Running Samples 6. Compiling OpenCL* Kernels 7. Intel® PAC with Intel® Arria® 10 GX FPGA Security for OpenCL* Applications 8. OpenCL* on the Intel® PAC with Intel® Arria® 10 GX FPGA Quick Start User Guide Archives 9. Document Revision History for OpenCL* Quick Start User Guide Intel® Programmable Acceleration Card with Intel® Arria® 10 GX FPGA
1. About this Document x
1.1. Conventions 1.2. Acceleration Glossary 1.3. Acronyms
2. Setting Up the Host Machine x
2.1. Introduction 2.2. Initializing the Environment for OpenCL* with Intel Acceleration Stack 2.3. Initializing the Intel® Acceleration Stack for OpenCL* in a Virtualized Environment
2.1. Introduction x
2.1.1. Release Content
5. Running Samples x
5.1. Running Hello World 5.2. Running Vector Add
6. Compiling OpenCL* Kernels x
6.1. Checking Timing Results
1. About this Document
2. Setting Up the Host Machine
3. Running Diagnostics
4. OpenCL* Support for Multi-Card Systems
5. Running Samples
6. Compiling OpenCL* Kernels
7. Intel® PAC with Intel® Arria® 10 GX FPGA Security for OpenCL* Applications
8. OpenCL* on the Intel® PAC with Intel® Arria® 10 GX FPGA Quick Start User Guide Archives
9. Document Revision History for OpenCL* Quick Start User Guide Intel® Programmable Acceleration Card with Intel® Arria® 10 GX FPGA

3. Running Diagnostics

Before running diagnostics, load an OpenCL* kernel to the board. The following instructions use the OpenCL* kernel or you may also use your own.

  1. Load OpenCL* kernel: 
    $ aocl program acl0 vector_add.aocx 
aocl program: Running program from /home/DCPsupport/intelrtestack/a10_gx_pac_ias_1_2_1_pv/opencl/opencl_bsp/linux64/libexec
Program succeed.
    Sample program output: 
    aocl program: Running program from $OPAE_PLATFORM_ROOT/opencl/opencl_bsp \
/linux64/libexec
Program succeed.
  2. Run the simple diagnostic utility: 
    $ aocl diagnose
    Sample diagnostic output: 
    --------------------------------------------------------------------
ICD System Diagnostics                                              
--------------------------------------------------------------------

Using the following location for ICD installation: 
	/etc/OpenCL/vendors

Found 1 icd entry at that location:
	/etc/OpenCL/vendors/Altera.icd

the following OpenCL libraries are referenced in the icd files:
	libalteracl.so

checking LD_LIBRARY_PATH for registered libraries:
	libalteracl.so was registered on the system at /<installation_directory>/tools/intelFPGA_pro/quartus_19.2.0b57/hld/host/linux64/lib

Using the following location for fcd installations:
	/opt/Intel/OpenCL/Boards

Found 1 fcd entry at that location:
	/opt/Intel/OpenCL/Boards/dcp_bsp.fcd

the following OpenCL libraries are referenced in the fcd files:
	/<installation_directory>/inteldevstack/a10_gx_pac_ias_1_2_1_pv/opencl/opencl_bsp/linux64/lib/libintel_opae_mmd.so

checking LD_LIBRARY_PATH for registered libraries:
	/<installation_directory>/inteldevstack/a10_gx_pac_ias_1_2_1_pv/opencl/opencl_bsp/linux64/lib/libintel_opae_mmd.so was registered on the system.

Number of Platforms = 1 
	1. Intel(R) FPGA SDK for OpenCL(TM)                             | Intel(R) Corporation           | OpenCL 1.0 Intel(R) FPGA SDK for OpenCL(TM), Version 19.4
--------------------------------------------------------------------
ICD diagnostics PASSED                                              
--------------------------------------------------------------------
--------------------------------------------------------------------
BSP Diagnostics                                                     
--------------------------------------------------------------------
--------------------------------------------------------------------
Device Name:
acl0

BSP Install Location:
/<installation_directory>/inteldevstack/a10_gx_pac_ias_1_2_1_pv/opencl/opencl_bsp

Vendor: Intel Corp

Physical Dev Name   Status            Information

pac_ef00000         Passed            Intel PAC Platform (pac_ef00000)
                                      PCIe 05:00.0
                                      FPGA temperature = 49 degrees C.

DIAGNOSTIC_PASSED
--------------------------------------------------------------------
  3. Run the advanced diagnostic: 
    $ aocl diagnose acl0
    Sample advanced diagnostic output: 
    --------------------------------------------------------------------
ICD System Diagnostics                                              
--------------------------------------------------------------------
Using the following location for ICD installation: 
	/etc/OpenCL/vendors

Found 1 icd entry at that location:
	/etc/OpenCL/vendors/Altera.icd

the following OpenCL libraries are referenced in the icd files:
	libalteracl.so

checking LD_LIBRARY_PATH for registered libraries:
	libalteracl.so was registered on the system at /storage/shared/home_directories/homichel/tools/intelFPGA_pro/quartus_19.2.0b57/hld/host/linux64/lib

Using the following location for fcd installations:
	/opt/Intel/OpenCL/Boards

Found 1 fcd entry at that location:
	/opt/Intel/OpenCL/Boards/dcp_bsp.fcd

the following OpenCL libraries are referenced in the fcd files:
	/storage/shared/home_directories/homichel/inteldevstack/a10_gx_pac_ias_1_2_1_pv/opencl/opencl_bsp/linux64/lib/libintel_opae_mmd.so

checking LD_LIBRARY_PATH for registered libraries:
	/storage/shared/home_directories/homichel/inteldevstack/a10_gx_pac_ias_1_2_1_pv/opencl/opencl_bsp/linux64/lib/libintel_opae_mmd.so was registered on the system.

Number of Platforms = 1 
	1. Intel(R) FPGA SDK for OpenCL(TM)                             | Intel(R) Corporation           | OpenCL 1.0 Intel(R) FPGA SDK for OpenCL(TM), Version 19.4
--------------------------------------------------------------------
ICD diagnostics PASSED                                              
--------------------------------------------------------------------
--------------------------------------------------------------------
BSP Diagnostics                                                     
--------------------------------------------------------------------
Using platform: Intel(R) FPGA SDK for OpenCL(TM)
Using Device with name: pac_a10 : Intel PAC Platform (pac_ef00000)
Using Device from vendor: Intel Corp
clGetDeviceInfo CL_DEVICE_GLOBAL_MEM_SIZE = 8589933568
clGetDeviceInfo CL_DEVICE_MAX_MEM_ALLOC_SIZE = 8589933568
Allocated 8589933568 bytes
Actual maximum buffer size = 8589933568 bytes
Writing 8191 MB to global memory ...
Allocated 1073741824 Bytes host buffer for large transfers
Write speed: 5843.54 MB/s [5689.33 -> 6183.29]
Reading and verifying 8191 MB from global memory ...
Read speed: 6663.87 MB/s [6523.73 -> 6729.59]
Successfully wrote and readback 8191 MB buffer

Transferring 262144 KBs in 512 512 KB blocks ... 3525.69 MB/s
Transferring 262144 KBs in 256 1024 KB blocks ... 3734.74 MB/s
Transferring 262144 KBs in 128 2048 KB blocks ... 3131.35 MB/s
Transferring 262144 KBs in 64 4096 KB blocks ... 4269.86 MB/s
Transferring 262144 KBs in 32 8192 KB blocks ... 4801.54 MB/s
Transferring 262144 KBs in 16 16384 KB blocks ... 5177.86 MB/s
Transferring 262144 KBs in 8 32768 KB blocks ... 5413.49 MB/s
Transferring 262144 KBs in 4 65536 KB blocks ... 5627.71 MB/s
Transferring 262144 KBs in 2 131072 KB blocks ... 6849.71 MB/s
Transferring 262144 KBs in 1 262144 KB blocks ... 6895.99 MB/s

As a reference:
PCIe Gen1 peak speed: 250MB/s/lane
PCIe Gen2 peak speed: 500MB/s/lane
PCIe Gen3 peak speed: 985MB/s/lane

Writing 262144 KBs with block size (in bytes) below:

Block_Size Avg    Max    Min    End-End (MB/s)
  524288 3073.65 3525.69 1750.25 2440.57
 1048576 2564.51 3734.74 1979.41 2208.89
 2097152 2437.80 2701.91 2240.24 2265.86
 4194304 3391.19 4220.46 2581.52 3236.38
 8388608 4225.16 4700.36 3138.08 4052.50
16777216 4419.41 4961.91 3436.53 4343.19
33554432 4775.78 5413.49 3895.03 4728.55
67108864 4889.22 5497.75 4160.56 4860.37
134217728 6233.47 6242.94 6224.03 6218.41
268435456 6218.76 6218.76 6218.76 6218.76

Reading 262144 KBs with block size (in bytes) below:

Block_Size Avg    Max    Min    End-End (MB/s)
  524288 3187.99 3457.18 2448.31 2631.75
 1048576 3077.55 3332.59 2491.45 2480.11
 2097152 2835.68 3131.35 1976.36 2540.81
 4194304 3675.78 4269.86 2552.75 3447.85
 8388608 4203.11 4801.54 2851.31 4056.30
16777216 4488.70 5177.86 3125.90 4398.89
33554432 4658.78 5391.93 3216.03 4608.17
67108864 4896.32 5627.71 3927.64 4870.05
134217728 6735.61 6849.71 6625.26 6726.29
268435456 6895.99 6895.99 6895.99 6895.99

Write top speed = 6242.94 MB/s
Read top speed = 6895.99 MB/s
Throughput = 6569.46 MB/s

DIAGNOSTIC_PASSED
Level Two Title