cancel
Showing results for 
Search instead for 
Did you mean: 

Archives Discussions

Meteorhead
Challenger

debugging kernel hang

Hi all!

My question is how should one go about debugging a kernel that tends to hang when given too much work? On a CPU I would suspect a memory leak, if hanging is work dependant, but is this possible at all on a GPU? No memory allocation is done dynamically, so I'd think this is a 'no way'.

I do not want to blame it on the compiler, the drivers or the hardware, I'm just curious what are the typical tactics of finding out what could be the cause of the problem?

The simulation is written in a manner, that every kernel call more and more work is given to the GPU, namely the for cycle processed inside gets longer. It is also lengthened by the size of area simulated, and there is a strong correllation between given work and tendency to hang.

How would you go about finding the problem?

0 Likes
6 Replies
himanshu_gautam
Grandmaster

meteorhead,

Are you working in windows or linux?

I think many people are actively using GDB to debug openCL kernels in linux.

0 Likes

I have read about the GDB tool under linux, but I have found it a little too un-userfriendly, but if you suggest that is the best way getting at it, I will try to make friends with it.

0 Likes
Lev
Journeyman III

Watch dog timer is not the reason for it?

0 Likes

Lev,

Can you explain your comment. I did not get it.

 

0 Likes

I mean that windows stop the kernell if it take too long time to run.

0 Likes

Lev,

Watchdog timer issue only comes in Windows. There also you can disable it using Registry settings.

 

0 Likes