6

I'm trying to use nvvp to profile opencl kernels. I'm running ubuntu 12.04 64b with a GTX 580 and have verified the CUDA toolkit is working fine (i can run and profile cuda code). When trying to debug my opencl code i get:
Warning: No CUDA application was profiled, exiting
Any hints?

Emanuel Ey
  • 2,724
  • 5
  • 30
  • 38
  • Possible duplicate of [Is there a way to profile an OpenCL or a pyOpenCL program?](http://stackoverflow.com/questions/29068229/is-there-a-way-to-profile-an-opencl-or-a-pyopencl-program) – Dschoni Feb 17 '16 at 13:21
  • See my answer on http://stackoverflow.com/questions/29068229/is-there-a-way-to-profile-an-opencl-or-a-pyopencl-program/35016313#35016313 – Dschoni Feb 17 '16 at 13:22

2 Answers2

5

Nvidia's visual profiler (nvvp) can be used to profile OpenCL programs, but it is more of a pain than profiling in CUDA directly.

Simon McIntosh's High Performance Computing group over at the University of Bristol came up with the original solution (here), and I can verify it works.

I'll summarise the basics:

  • Firstly, the environment variable COMPUTE_PROFILE must be set, this is done with COMPUTE_PROFILE=1

  • Secondly a COMPUTE_PROFILE_CONFIG must be provided, a sample I use (called nvvp.cfg) contains:

    profilelogformat CSV
streamid
gpustarttimestamp
gpuendtimestamp

  • Next to perform the actual profiling, in this case I'll profile an OpenCL application called HuffFramework, using:

    COMPUTE_PROFILE=1 COMPUTE_PROFILE_CONFIG=nvvp.cfg ./HuffFramework

  • This then generates a series of opencl_profile_*.log files, where * is the number of threads.

  • These log files can't be loaded by nvvp just yet as all kernel function symbols have a leading OPENCL_ instead of an expected CUDA_, thus replace these symbols with a quick script like so:

    sed 's/OPENCL_/CUDA_/g' opencl_profile_0.log > cuda_profile_0.log

  • Finally cuda_profile_0.log can now be imported by nvvp, by starting nvvp and going File->Import...->Command-line Profiler, point it to cuda_profile_0.log and preso!

enter image description here

Beau Johnston
  • 116
  • 1
  • 4
0

nvvp can only profile CUDA applications.

Eugene
  • 9,242
  • 2
  • 30
  • 29
  • I feared as much, although i can find no mention of this anywhere in NVidia's documentation... Is there an alternative? – Emanuel Ey Aug 23 '12 at 15:55
  • 1
    Although it is still an under development project, you can obtain a timeline view of your OpenCL application using LTPV: http://code.google.com/p/ltpv/ – simon.denel Sep 06 '13 at 12:11
  • Actually, this is not 100% true. http://uob-hpc.github.io/2015/05/27/nvvp-import-opencl/ – Dschoni Feb 17 '16 at 13:20
  • It's now here: http://uob-hpc.github.io/2015/05/27/nvvp-import-opencl.html – Lars Oct 19 '17 at 10:15