Community
cancel
Showing results for 
Search instead for 
Did you mean: 
KG0001
Beginner
233 Views

Can't access FPGA machine via Intel Dev Cloud

Hi Intel Support Team, On Intel Dev Cloud machine, I could able to build my sample program for default device selector (GPU device), that works fine. But if try to compile the same sample vector-add program for FPGA hardware by using 'make hw -f Makefile.fpga' makefile command, getting error. It seems the issue is specific to my Intel dev cloud login. The FPGA device not able to find for my Dev Cloud login, the '‌aocl diagnose' command also failed. I attached that build and aocl log here for you reference. Note: I am trying this intel provided sample program only. https://github.com/intel/BaseKit-code-samples/tree/master/DPC%2B%2BCompiler/vector-add Regards, Kalyan-G.
Tags (1)
0 Kudos
6 Replies
GouthamK_Intel
Moderator
233 Views

Hi Kalyan,
Thanks for reaching out to us!

On Devcloud, we are having dedicated compute nodes which are having FPGA hardware. As per the logs provided by you, we observed that the node which you are running your application contains only an FPGA Emulation device but does not contain FPGA hardware.

We suggest you acquire an FPGA hardware node via an interactive terminal or submit a job to a node that has FPGA hardware.

 

Example of Job submission for FPGA hardware node. 

qsub -l nodes=1:fpga_runtime:ppn=2,walltime=08:00:00 -d . build_fpga_hw.sh (arg Small L(-l)).

Example of an interactive terminal

qsub -I -l nodes=1:fpga_runtime:ppn=2  (1st arg: Captial i (-I), 2nd arg: Small L(-l)).

 

If you want to compile and run your application on the FPGA Emulation device, use the below commands.

Example of Job submission for FPGA Emulation node. 

qsub -l nodes=1:fpga_compile:ppn=2,walltime=08:00:00 -d . myscript.sh  (arg Small L(-l)).

Example of an interactive terminal

qsub -I -l nodes=1:fpga_compile:ppn=2  (1st arg: Captial i (-I), 2nd arg: Small L(-l)).

 

As FPGA issues are not handled here. If you have any issues related to FPGA, please ask your question in the Intel® High Level Design Forum.

 

Regards

Goutham

 

KG0001
Beginner
233 Views

Thanks Goutham.
I am using below command which causing this problem. Let me change this command as you suggest. 

qsub -l nodes=1:gpu:ppn=2 -d . $script

Regards,

Kalyan-G.

 

GouthamK_Intel
Moderator
233 Views

Hi Kalyan,

Could you please let us know if your issue still persists.

If it is resolved, please do let us know if we can close this thread.

 

Thanks & Regards

Goutham

GouthamK_Intel
Moderator
233 Views

Hi Kalyan,

Could you please let us know whether we can close this thread if your issue is resolved?

If your issue still persists, please do let us know. So that we will be able to help you with that. 

 

 

Thanks

Goutham

KG0001
Beginner
233 Views

Hi Goutham,

Apology for delayed reply. Your suggestion works good. now i could able to access the FPGA device by using 'qsub -l nodes=1:fpga_runtime:ppn=2' node. thanks for your support.

please close this ticket. 

Regards,

Kalyan-G 

GouthamK_Intel
Moderator
233 Views

Hi Kalyan,

Glad to hear that solution provided helped you.!

We are closing this thread.

Please feel free to raise new threads in case of any further issues. 

Have a Good day!

 

Regards

Goutham

Reply