R3 2019 Error : FAILED to read auto-discovery string at byte 44

[Rushikesh_P_Intel]

Hi,

   I have Arria 10 PAC card with Stack 1.2 and OpenVINO 2019R3 

I am gettting the below error , please help to find the root cause

FAILED to read auto-discovery string at byte 44. Full auto-discovery string value is 19 f22d523feb5f381fd20cb28240ee547b4686c3eb pac_a10 0 0 2 4294967296 0 4294967296 4294967296 8589934592 0 9 DLA_ARCH_8x32_fp11_sb12768_i1_actk8_poolk8_max_window6_owk4_image224x224x4096_elu8_mbfr 0 128 1 0 0 1 0 1 0 0 0 0 1 1 1 3 1 1 1 0 1 DLA_CHECKSUM_44b1ad592daeb99b 256 128 1 0 0 1 0 1 0 0 0 0 1 1 1 3 1 1 1 0 1 DLA_VERSION_2019R2 512 128 1 0 0 1 0 1 0 0 0 0 1 1 1 3 1 1 1 0 1 bias_reader 768 256 1 0 0 1 0 1 0 2 2 1 8 2 1 8 0 0 1 1 1 3 1 1 1 0 1 config_reader 1024 256 1 0 0 1 0 1 0 2 2 1 8 0 0 4 0 0 1 1 1 3 1 1 1 0 1 filter_reader 1280 256 1 0 0 1 0 1 0 3 2 1 8 2 1 8 0 0 4 0 0 1 1 1 3 1 1 1 0 1 input_reader 1536 256 1 0 0 1 0 1 0 4 2 1 8 2 1 8 0 0 4 0 0 4 0 0 1 1 1 3 1 1 1 0 1 output_writer 1792 256 1 0 0 1 0 1 0 2 2 1 8 2 1 8 0 0 1 1 1 3 1 1 1 0 1 perf_writer 2048 256 1 0 0 1 0 1 0 1 2 1 8 0 0 1 1 1 3 1 1 1 0 1
acl_hal_mmd.c:1393:assert failure: Failed to initialize kernel interfaceclassification_sample_async: acl_hal_mmd.c:1393: l_try_device: Assertion `0' failed.
Aborted (core dumped)
 

Regards,

Rushikesh

[Max_L_Intel]

Hi Rushikesh.

It seems the card has not been detected properly. Please ensure that you have followed all the steps correctly from the below links:

1. OpenVINO Installation Guide for FPGA - https://docs.openvinotoolkit.org/2019_R3.1/_docs_install_guides_installing_openvino_linux_fpga.html

2. OpenVINO Toolkit FPGA Configuration Guide, and especially the section "Verify the Intel® Acceleration Stack for FPGAs OpenCL™ BSP" in it - https://docs.openvinotoolkit.org/2019_R3.1/_docs_install_guides_PAC_Configure_2019RX.html

Previously, we've seen a similar issue fixed by correct setting the variable INTELFPGAOCLSDKROOT within init_openvino.sh that you need to create.

Best regards, Max,

[Rushikesh_P_Intel]

I resolved that issue , thanks 

I am getting one other issue , the error is same with HETRO or CPU only 

regex Error 

InferenceEngine: 0x7f4ec32a7030
[ INFO ] Parsing input parameters
[ INFO ] Reading input
[ INFO ] Use async mode for Face Detection
[ INFO ] Use async mode for Age/Gender
[ INFO ] Use async mode for Head Pose
[ INFO ] Use async mode for Emotions Recognition
[ INFO ] Use async mode for Facial Landmarks
[ INFO ] Loading device CPU
        CPU
        MKLDNNPlugin version ......... 2.1
        Build ........... 37988

[ INFO ] Loading network files for Face Detection
[ INFO ] Batch size is set to 1
[ INFO ] Checking Face Detection network inputs
[ INFO ] Checking Face Detection network outputs
[ INFO ] Loading Face Detection model to the CPU device
[ INFO ] Loading network files for Age/Gender Recognition network
[ INFO ] Batch size is set to 16 for Age/Gender Recognition network
[ INFO ] Checking Age/Gender Recognition network inputs
[ INFO ] Checking Age/Gender Recognition network outputs
[ INFO ] Age layer: age_conv3
[ INFO ] Gender layer: prob
[ INFO ] Loading Age/Gender Recognition model to the CPU plugin
[ INFO ] Loading network files for Head Pose Estimation network
[ INFO ] Batch size is set to  16 for Head Pose Estimation network
[ INFO ] Checking Head Pose Estimation network inputs
[ INFO ] Checking Head Pose Estimation network outputs
[ INFO ] Loading Head Pose Estimation model to the CPU plugin
[ INFO ] Loading network files for Emotions Recognition
[ INFO ] Batch size is set to 16 for Emotions Recognition
[ INFO ] Checking Emotions Recognition network inputs
[ INFO ] Checking Emotions Recognition network outputs
[ INFO ] Emotions layer: prob_emotion
[ INFO ] Loading Emotions Recognition model to the CPU plugin
[ INFO ] Loading network files for Facial Landmarks Estimation
[ INFO ] Batch size is set to  16 for Facial Landmarks Estimation network
[ INFO ] Checking Facial Landmarks Estimation network inputs
[ INFO ] Checking Facial Landmarks Estimation network outputs
[ INFO ] Loading Facial Landmarks Estimation model to the CPU plugin
[ INFO ] Start inference
To close the application, press 'CTRL+C' here or switch to the output window and press Q or Esc
[ ERROR ] regex_error
 

Regards,

Rushikesh

[Max_L_Intel]

Hi Rushikesh.

This error might be related to many multiple factors. To investigate this further I need some more information from you.

1) What gcc version you have installed?

To check current gcc version installed: gcc --version
To update gcc to latest version: sudo apt-get install gcc

2) Do you see the same error when you run this demo with FPGA plugin only and with one model only (Face Detection)?

3) Are you able to run "demo_security_barrier_camera.sh" using CPU plugin only? From here https://docs.openvinotoolkit.org/latest/_docs_install_guides_installing_openvino_linux_fpga.html

And also classification_sample_async using "-d HETERO:FPGA,CPU" from here https://docs.openvinotoolkit.org/latest/_docs_install_guides_PAC_Configure_2019RX.html

4) If this issue still remains, please let us know what exact application (is it interactive_face_detection_demo?) and what input -i video file or cam device you try to utilize here?

Best regards, Max.

[Rushikesh_P_Intel]

I am using Centos 7.4

gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39) , its Latest

demo_security_barrier_camera.sh also gives the error

Run Inference Engine security_barrier_camera demo

Run ./security_barrier_camera_demo -d CPU -d_va CPU -d_lpr CPU -i /opt/intel/openvino_2020.1.023/deployment_tools/demo/car_1.bmp -m /root/openvino_models/ir/intel/vehicle-license-plate-detection-barrier-0106/FP16/vehicle-license-plate-detection-barrier-0106.xml -m_lpr /root/openvino_models/ir/intel/license-plate-recognition-barrier-0001/FP16/license-plate-recognition-barrier-0001.xml -m_va /root/openvino_models/ir/intel/vehicle-attributes-recognition-barrier-0039/FP16/vehicle-attributes-recognition-barrier-0039.xml

[ INFO ] InferenceEngine: 0x7fd2dc278030
[ INFO ] Files were added: 1
[ INFO ]     /opt/intel/openvino_2020.1.023/deployment_tools/demo/car_1.bmp
[ INFO ] Loading device CPU
        CPU
        MKLDNNPlugin version ......... 2.1
        Build ........... 37988

[ INFO ] Loading detection model to the CPU plugin
[ INFO ] Loading Vehicle Attribs model to the CPU plugin
[ INFO ] Loading Licence Plate Recognition (LPR) model to the CPU plugin
[ INFO ] Number of InferRequests: 1 (detection), 3 (classification), 3 (recognition)
[ INFO ] 6 streams for CPU
[ INFO ] Display resolution: 1920x1080
[ ERROR ] regex_error
Error on or near line 15; exiting with status 1
 

I am trying to run the interactive face detection DEMO , i just want to make sure that openVINO works fine.

It was working , i had to format this machine and then i installed it with openvino_2020.1.023 , earlier it was R3

[Rushikesh_P_Intel]

Mark ,

classification_sample_async using "-d HETERO:FPGA,CPU" 

This works fine with HETERO as well as CPU only options 

I think is it because i am accessing this Machine remotely , i am using the SSH

I dont expect the Video to run in the SSH , but what i expect is that it should stop at 

Start Inference 

example :

[ INFO ] Checking Facial Landmarks Estimation network inputs
[ INFO ] Checking Facial Landmarks Estimation network outputs
[ INFO ] Loading Facial Landmarks Estimation model to the CPU plugin
[ INFO ] Start inference

but in this case its throwing error , this was not the case with R3.

is it because in both the example interactive face detection and security barrier demo , it needs to pop-up a video/image it does not like it ?

As i said in R3 :-

i was able to run the Security Demo via SSH 

Interactive Face demo use to run till Start inference and then stop ( which is expected )

[Rushikesh_P_Intel]

Its same error even when i run the DEMO directly from the server machine.

[Max_L_Intel]

Hi Rushikesh.

is it because in both the example interactive face detection and security barrier demo , it needs to pop-up a video/image it does not like it ?

Yes, both interactive_face_detection and security_barrier_camera applications require GUI mode to run. On my end I've just tested it on 2019.R3 and 2020.1 builds on Ubuntu 18.04, and I see the following error for both of them in CLI mode:

[ INFO ] InferenceEngine: 
    API version ............ 2.1
    Build .................. 37988
    Description ....... API
[ INFO ] Files were added: 1
[ INFO ]     /opt/intel/openvino_2019.3.376/deployment_tools/demo/car.png
[ INFO ] Loading device CPU
    CPU
    MKLDNNPlugin version ......... 2.1
    Build ........... 37988

[ INFO ] Loading detection model to the CPU plugin
[ INFO ] Number of InferRequests: 1 (detection), 0 (classification), 0 (recognition)
[ INFO ] 4 streams for CPU
[ INFO ] Display resolution: 1920x1080
[ INFO ] Number of allocated frames: 3
[ INFO ] Resizable input with support of ROI crop and auto resize is disabled
Error on or near line 198; exiting with status 1
Unable to init server: Could not connect: Connection refused
(Detection results:1452): Gtk-WARNING **: 12:19:00.322: cannot open display

However, applications that don't input or output any graphics directly on camera or screen (like classification_sample_async or demo_squeezenet_download_convert_run) work fine in CLI mode. Please try to run them as well in order to ensure OpenVINO Toolkit being configured correctly.

Best regards, Max.

[Rushikesh_P_Intel]

Hi Max,

i tried to run on the server directly , but still same error.

I formatted the system and re-install again in ubuntu 

OpenVino R3 

This time i am getting different error , Installation process is really not very robust 

[ ERROR ] Failed to create plugin /opt/intel/openvino_2019.3.376/deployment_tools/inference_engine/lib/intel64/libdliaPlugin.so for device FPGA

Please, check your environment

Cannot load library '/opt/intel/openvino_2019.3.376/deployment_tools/inference_engine/lib/intel64/libdliaPlugin.so': libalteracl.so: cannot open shared object file: No such file or directory

Can you point out whats missing ?

Script file :

source /home/intel/intelrtestack/init_env.sh
export CL_CONTEXT_COMPILER_MODE_ALTERA=3
export CL_CONTEXT_COMPILER_MODE_INTELFPGA=3
export INTELFPGAOCLSDKROOT="/opt/altera/aocl-pro-rte/aclrte-linux64/"
export ALTERAOCLSDKROOT="\$INTELFPGAOCLSDKROOT"
export AOCL_BOARD_PACKAGE_ROOT="/home/intel/intelrtestack/a10_gx_pac_ias_1_2_pv/opencl/opencl_bsp/"
source $AOCL_BOARD_PACKAGE_ROOT/linux64/libexec/setup_permissions.sh
source $INTELFPGAOCLSDKROOT/init_opencl.sh
export IE_INSTALL="/opt/intel/openvino_2019.3.376/deployment_tools/"
source /opt/intel/openvino_2019.3.376/bin/setupvars.sh
export PATH="\$PATH:/home/intel/omz_demos_build/intel64/Release"
alias mo="python3.6 \$IE_INSTALL/model_optimizer/mo.py"

[Max_L_Intel]

Hi Rushikesh.

Have you installed a special FPGA supported OpenVINO Toolkit package? https://software.intel.com/en-us/openvino-toolkit/choose-download/free-download-linux-fpga

Package name: l_openvino_toolkit_fpga_p_2019.3.376
And once it is installed, an installation directory name should be something like openvino_fpga_2019.3.376/

If you have this FPGA supported OpenVINO toolkit package installed, then please try to locate a required libdliaPlugin.so file with

sudo find / -name '*libdliaPlugin.so'

[dangu]

Rushikesh,

I know this is from a while back, but any chance you remember what specifically resolved your original issue running the squeezenet demo (FAILED to read auto-discovery string at byte 44)? I am running into this same exact issue with 2020.2 and haven't had any luck repeating the steps from the Installation Guide or Configuration Guide. (I've also got the Arria 10 PAC)

My INTELFPGAOCLSDKROOT is set to ~/intelrtestack/opencl_rte/aclrte-linux64

If anyone has an idea about the root cause please let me know. Thanks!

[Micha_M_]

I can't download OpenVino version 2020.2 for windows.  My openVino license has expired. 
How to extend the licence for OpenVino?
Is this version free?
 

[Max_L_Intel]

Micha M. wrote:

I can't download OpenVino version 2020.2 for windows.  My openVino license has expired. 
How to extend the licence for OpenVino?
Is this version free?
 

Yes, it is free. Try this link https://software.seek.intel.com/openvino-toolkit?os=windows
You have to register there or log in.

[Rushikesh_P_Intel]

dangu wrote:

Rushikesh,

I know this is from a while back, but any chance you remember what specifically resolved your original issue running the squeezenet demo (FAILED to read auto-discovery string at byte 44)? I am running into this same exact issue with 2020.2 and haven't had any luck repeating the steps from the Installation Guide or Configuration Guide. (I've also got the Arria 10 PAC)

My INTELFPGAOCLSDKROOT is set to ~/intelrtestack/opencl_rte/aclrte-linux64

If anyone has an idea about the root cause please let me know. Thanks!

I moved to ubuntu 18.04 and re-installed it.